ICPR2020 Paper Browser

Paper download is intended for registered attendees only, and is subjected to the IEEE Copyright Policy. Any other use is strongly forbidden.

On Resource-Efficient Bayesian Network Classifiers and Deep Neural Networks

Wolfgang Roth, Günther Schindler, Holger Fröning, Franz Pernkopf

Auto-TLDR; Quantization-Aware Bayesian Network Classifiers for Small-Scale Scenarios

Abstract Slides Poster

We present two methods to reduce the complexity of Bayesian network (BN) classifiers. First, we introduce quantization-aware training using the straight-through gradient estimator to quantize the parameters of BNs to few bits. Second, we extend a recently proposed differentiable tree-augmented naive Bayes (TAN) structure learning approach to also consider the model size. Both methods are motivated by recent developments in the deep learning community, and they provide effective means to trade off between model size and prediction accuracy, which is demonstrated in extensive experiments. Furthermore, we contrast quantized BN classifiers with quantized deep neural networks (DNNs) for small-scale scenarios which have hardly been investigated in the literature. We show Pareto optimal models with respect to model size, number of operations, and test error and find that both model classes are viable options.

Similar papers

Resource-efficient DNNs for Keyword Spotting using Neural Architecture Search and Quantization

David Peter, Wolfgang Roth, Franz Pernkopf

Auto-TLDR; Neural Architecture Search for Keyword Spotting in Limited Resource Environments

Abstract Slides Poster Similar

This paper introduces neural architecture search (NAS) for the automatic discovery of small models for keyword spotting (KWS) in limited resource environments. We employ a differentiable NAS approach to optimize the structure of convolutional neural networks (CNNs) to meet certain memory constraints for storing weights as well as constraints on the number of operations per inference. Using NAS only, we were able to obtain a highly efficient model with 95.6% accuracy on the Google speech commands dataset with 494.8 kB of memory usage and 19.6 million operations. Additionally, weight quantization is used to reduce the memory consumption even further. We show that weight quantization to low bit-widths (e.g. 1 bit) can be used without substantial loss in accuracy. By increasing the number of input features from 10 MFCC to 20 MFCC we were able to increase the accuracy to 96.6% at 340.1 kB of memory usage and 27.1 million operations.

E-DNAS: Differentiable Neural Architecture Search for Embedded Systems

Javier García López, Antonio Agudo, Francesc Moreno-Noguer

Auto-TLDR; E-DNAS: Differentiable Architecture Search for Light-Weight Networks for Image Classification

Abstract Slides Poster Similar

Designing optimal and light weight networks to fit in resource-limited platforms like mobiles, DSPs or GPUs is a challenging problem with a wide range of interesting applications, {\em e.g.} in embedded systems for autonomous driving. While most approaches are based on manual hyperparameter tuning, there exist a new line of research, the so-called NAS (Neural Architecture Search) methods, that aim to optimize several metrics during the design process, including memory requirements of the network, number of FLOPs, number of MACs (Multiply-ACcumulate operations) or inference latency. However, while NAS methods have shown very promising results, they are still significantly time and cost consuming. In this work we introduce E-DNAS, a differentiable architecture search method, which improves the efficiency of NAS methods in designing light-weight networks for the task of image classification. Concretely, E-DNAS computes, in a differentiable manner, the optimal size of a number of meta-kernels that capture patterns of the input data at different resolutions. We also leverage on the additive property of convolution operations to merge several kernels with different compatible sizes into a single one, reducing thus the number of operations and the time required to estimate the optimal configuration. We evaluate our approach on several datasets to perform classification. We report results in terms of the SoC (System on Chips) metric, typically used in the Texas Instruments TDA2x families for autonomous driving applications. The results show that our approach allows designing low latency architectures significantly faster than state-of-the-art.

Generalization Comparison of Deep Neural Networks Via Output Sensitivity

Mahsa Forouzesh, Farnood Salehi, Patrick Thiran

Auto-TLDR; Generalization of Deep Neural Networks using Sensitivity

Abstract Slides Similar

Although recent works have brought some insights into the performance improvement of techniques used in state-of-the-art deep-learning models, more work is needed to understand their generalization properties. We shed light on this matter by linking the loss function to the output's sensitivity to its input. We find a rather strong empirical relation between the output sensitivity and the variance in the bias-variance decomposition of the loss function, which hints on using sensitivity as a metric for comparing the generalization performance of networks, without requiring labeled data. We find that sensitivity is decreased by applying popular methods which improve the generalization performance of the model, such as (1) using a deep network rather than a wide one, (2) adding convolutional layers to baseline classifiers instead of adding fully-connected layers, (3) using batch normalization, dropout and max-pooling, and (4) applying parameter initialization techniques.

Compression Strategies and Space-Conscious Representations for Deep Neural Networks

Giosuè Marinò, Gregorio Ghidoli, Marco Frasca, Dario Malchiodi

Auto-TLDR; Compression of Large Convolutional Neural Networks by Weight Pruning and Quantization

Abstract Slides Poster Similar

Recent advances in deep learning have made available large, powerful convolutional neural networks (CNN) with state-of-the-art performance in several real-world applications. Unfortunately, these large-sized models have millions of parameters, thus they are not deployable on resource-limited platforms (e.g. where RAM is limited). Compression of CNNs thereby becomes a critical problem to achieve memory-efficient and possibly computationally faster model representations. In this paper, we investigate the impact of lossy compression of CNNs by weight pruning and quantization, and lossless weight matrix representations based on source coding. We tested several combinations of these techniques on four benchmark datasets for classification and regression problems, achieving compression rates up to 165 times, while preserving or improving the model performance.

Revisiting the Training of Very Deep Neural Networks without Skip Connections

Oyebade Kayode Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Bjorn Ottersten

Auto-TLDR; Optimization of Very Deep PlainNets without shortcut connections with 'vanishing and exploding units' activations'

Abstract Slides Poster Similar

Deep neural networks (DNNs) with many layers of feature representations yield state-of-the-art results on several difficult learning tasks. However, optimizing very deep DNNs without shortcut connections known as PlainNets, is a notoriously hard problem. Considering the growing interest in this area, this paper investigates holistically two scenarios that plague the training of very deep PlainNets: (1) the relatively popular challenge of 'vanishing and exploding units' activations', and (2) the less investigated 'singularity' problem, which is studied in details in this paper. In contrast to earlier works that study only the saturation and explosion of units' activations in isolation, this paper harmonizes the inconspicuous coexistence of the aforementioned problems for very deep PlainNets. Particularly, we argue that the aforementioned problems would have to be tackled simultaneously for the successful training of very deep PlainNets. Finally, different techniques that can be employed for tackling the optimization problem are discussed, and a specific combination of simple techniques that allows the successful training of PlainNets having up to 100 layers is demonstrated.

VPU Specific CNNs through Neural Architecture Search

Ciarán Donegan, Hamza Yous, Saksham Sinha, Jonathan Byrne

Auto-TLDR; Efficient Convolutional Neural Networks for Edge Devices using Neural Architecture Search

Abstract Slides Poster Similar

The success of deep learning at computer vision tasks has led to an ever-increasing number of applications on edge devices. Often with the use of edge AI hardware accelerators like the Intel Movidius Vision Processing Unit (VPU). Performing computer vision tasks on edge devices is challenging. Many Convolutional Neural Networks (CNNs) are too complex to run on edge devices with limited computing power. This has created large interest in designing efficient CNNs and one promising way of doing this is through Neural Architecture Search (NAS). NAS aims to automate the design of neural networks. NAS can also optimize multiple different objectives together, like accuracy and efficiency, which is difficult for humans. In this paper, we use a differentiable NAS method to find efficient CNNs for VPU that achieves state-of-the-art classification accuracy on ImageNet. Our NAS designed model outperforms MobileNetV2, having almost 1\% higher top-1 accuracy while being 13\% faster on MyriadX VPU. To the best of our knowledge, this is the first time a VPU specific CNN has been designed using a NAS algorithm. Our results also reiterate the fact that efficient networks must be designed for each specific hardware. We show that efficient networks targeted at different devices do not perform as well on the VPU.

Towards Low-Bit Quantization of Deep Neural Networks with Limited Data

Yong Yuan, Chen Chen, Xiyuan Hu, Silong Peng

Auto-TLDR; Low-Precision Quantization of Deep Neural Networks with Limited Data

Abstract Slides Poster Similar

Recent machine learning methods use increasingly large deep neural networks to achieve state-of-the-art results in various tasks. Network quantization can effectively reduce computation and memory costs without modifying network structures, facilitating the deployment of deep neural networks (DNNs) on cloud and edge devices. However, most of the existing methods usually need time-consuming training or fine-tuning and access to the original training dataset that may be unavailable due to privacy or security concerns. In this paper, we present a novel method to achieve low-precision quantization of deep neural networks with limited data. Firstly, to reduce the complexity of per-channel quantization and the degeneration of per-layer quantization, we introduce group-wise quantization which separates the output channels into groups that each group is quantized separately. Secondly, to better distill knowledge from the pre-trained FP32 model with limited data, we introduce a two-stage knowledge distillation method that divides the optimization process into independent optimization stage and joint optimization stage to address the limitation of layer-wise supervision and global supervision. Extensive experiments on ImageNet2012 (ResNet18/50, ShuffleNetV2, and MobileNetV2) demonstrate that the proposed approach can significantly improve the quantization model's accuracy when only a few training samples are available. We further show that the method also extends to other computer vision architectures and tasks such as object detection and semantic segmentation.

Fast Implementation of 4-Bit Convolutional Neural Networks for Mobile Devices

Anton Trusov, Elena Limonova, Dmitry Slugin, Dmitry Nikolaev, Vladimir V. Arlazarov

Auto-TLDR; Efficient Quantized Low-Precision Neural Networks for Mobile Devices

Abstract Slides Poster Similar

Quantized low-precision neural networks are very popular because they require less computational resources for inference and can provide high performance, which is vital for real-time and embedded recognition systems. However, their advantages are apparent for FPGA and ASIC devices, while general-purpose processor architectures are not always able to perform low-bit integer computations efficiently. The most frequently used low-precision neural network model for mobile central processors is an 8-bit quantized network. However, in a number of cases, it is possible to use fewer bits for weights and activations, and the only problem is the difficulty of efficient implementation. We introduce an efficient implementation of 4-bit matrix multiplication for quantized neural networks and perform time measurements on a mobile ARM processor. It shows 2.9 times speedup compared to standard floating-point multiplication and is 1.5 times faster than 8-bit quantized one. We also demonstrate a 4-bit quantized neural network for OCR recognition on the MIDV-500 dataset. 4-bit quantization gives 95.0% accuracy and 48% overall inference speedup, while an 8-bit quantized network gives 95.4% accuracy and 39% speedup. The results show that 4-bit quantization perfectly suits mobile devices, yielding good enough accuracy and low inference time.

ResNet-Like Architecture with Low Hardware Requirements

Elena Limonova, Daniil Alfonso, Dmitry Nikolaev, Vladimir V. Arlazarov

Auto-TLDR; BM-ResNet: Bipolar Morphological ResNet for Image Classification

Abstract Slides Poster Similar

One of the most computationally intensive parts in modern recognition systems is an inference of deep neural networks that are used for image classification, segmentation, enhancement, and recognition. The growing popularity of edge computing makes us look for ways to reduce its time for mobile and embedded devices. One way to decrease the neural network inference time is to modify a neuron model to make it more efficient for computations on a specific device. The example of such a model is a bipolar morphological neuron model. The bipolar morphological neuron is based on the idea of replacing multiplication with addition and maximum operations. This model has been demonstrated for simple image classification with LeNet-like architectures [1]. In the paper, we introduce a bipolar morphological ResNet (BM-ResNet) model obtained from a much more complex ResNet architecture by converting its layers to bipolar morphological ones. We apply BM-ResNet to image classification on MNIST and CIFAR-10 datasets with only a moderate accuracy decrease from 99.3% to 99.1% and from 85.3% to 85.1%. We also estimate the computational complexity of the resulting model. We show that for the majority of ResNet layers, the considered model requires 2.1-2.9 times fewer logic gates for implementation and 15-30% lower latency.

A Discriminant Information Approach to Deep Neural Network Pruning

Zejiang Hou, Sy Kung

Auto-TLDR; Channel Pruning Using Discriminant Information and Reinforcement Learning

Abstract Slides Poster Similar

Network pruning has become the de facto tool to accelerate and compress deep convolutional neural networks for mobile and edge applications. Previous works tend to perform channel selection in layer-wise manner based on predefined heuristics, without considering layer importance or systematically optimizing the pruned structure. In this work, we propose a novel channel pruning method that jointly harnesses two strategies: (1) a channel importance ranking heuristics based on the feature-maps discriminant power, (2) a searching method for optimal pruning budget allocation. For the former, we propose a Discriminant Information (DI) based channel selection algorithm. We use a small batch of training samples to compute the DI score for each channel and rank the channel importance so that channels really contributing to the feature-maps discriminant power are retained. For the latter, in order to search the optimal pruning budget allocation, we formulate a reward maximization problem to discover the layer importance and generating the pruning budget accordingly. Such reward maximization can be efficiently solved by the policy gradient algorithm in reinforcement learning, yielding our final pruned network which achieves the best accuracy-efficiency trade-off. Experiments on a variety of CNN architectures and benchmark datasets show that our proposed channel pruning methods compare favorably with previous state-of-the-art methods. On ImageNet, our pruned MobileNetV2 outperforms the previous layer-wise state-of-the-art pruning method CPLI \cite{guo2020channel} by 2\% Top-1 accuracy while reducing the FLOPs by 50\%.

Filtered Batch Normalization

András Horváth, Jalal Al-Afandi

Auto-TLDR; Batch Normalization with Out-of-Distribution Activations in Deep Neural Networks

Abstract Slides Poster Similar

It is a common assumption that the activation of different layers in neural networks follow Gaussian distribution. This distribution can be transformed using normalization techniques, such as batch-normalization, increasing convergence speed and improving accuracy. In this paper we would like to demonstrate, that activations do not necessarily follow Gaussian distribution in all layers. Neurons in deeper layers are more and more specific which can result extremely large, out-of-distribution activations. We will demonstrate that one can create more consistent mean and variance values for batch normalization during training by filtering out these activations which can further improve convergence speed and yield higher validation accuracy.

Learning with Multiplicative Perturbations

Xiulong Yang, Shihao Ji

Auto-TLDR; XAT and xVAT: A Multiplicative Adversarial Training Algorithm for Robust DNN Training

Abstract Slides Poster Similar

Adversarial Training (AT) and Virtual Adversarial Training (VAT) are the regularization techniques that train Deep Neural Networks (DNNs) with adversarial examples generated by adding small but worst-case perturbations to input examples. In this paper, we propose xAT and xVAT, new adversarial training algorithms that generate multiplicative perturbations to input examples for robust training of DNNs. Such perturbations are much more perceptible and interpretable than their additive counterparts exploited by AT and VAT. Furthermore, the multiplicative perturbations can be generated transductively or inductively, while the standard AT and VAT only support a transductive implementation. We conduct a series of experiments that analyze the behavior of the multiplicative perturbations and demonstrate that xAT and xVAT match or outperform state-of-the-art classification accuracies across multiple established benchmarks while being about 30% faster than their additive counterparts. Our source code can be found at https://github.com/sndnyang/xvat

Operation and Topology Aware Fast Differentiable Architecture Search

Shahid Siddiqui, Christos Kyrkou, Theocharis Theocharides

Auto-TLDR; EDARTS: Efficient Differentiable Architecture Search with Efficient Optimization

Abstract Slides Poster Similar

Differentiable architecture search (DARTS) has gained significant attention amongst neural architecture search approaches due to its effectiveness in finding competitive network architectures with reasonable computational complexity. DARTS' search space however is designed such that even a randomly picked architecture is very competitive and due to the complexity of search architectural building block or cell, it is unclear whether these are certain operations or the cell topology that contributes most to achieving higher final accuracy. In this work, we dissect the DARTS's search space as to understand which components are most effective in producing better architectures. Our experiments show that: (1) Good architectures can be found regardless of the search network depth; (2) Seperable convolution is the most effective operation in the search space; and (3) The cell topology also has substantial effect on the accuracy. Based on these insights, we propose an efficient search approach based referred to as eDARTS, that searches on a pre-specified cell with a good topology with increased attention to important operations using a shallow supernet. Moreover, we propose some optimizations for eDARTS which significantly speed up the search as well as alleviate the well known skip connection aggregation problem of DARTS. eDARTS achieves an error rate of 2.53% on CIFAR-10 using a 3.1M parameters model; while the search cost is less than 30 minutes.

On the Information of Feature Maps and Pruning of Deep Neural Networks

Mohammadreza Soltani, Suya Wu, Jie Ding, Robert Ravier, Vahid Tarokh

Auto-TLDR; Compressing Deep Neural Models Using Mutual Information

Abstract Slides Poster Similar

A technique for compressing deep neural models achieving competitive performance to state-of-the-art methods is proposed. The approach utilizes the mutual information between the feature maps and the output of the model in order to prune the redundant layers of the network. Extensive numerical experiments on both CIFAR-10, CIFAR-100, and Tiny ImageNet data sets demonstrate that the proposed method can be effective in compressing deep models, both in terms of the numbers of parameters and operations. For instance, by applying the proposed approach to DenseNet model with 0.77 million parameters and 293 million operations for classification of CIFAR-10 data set, a reduction of 62.66% and 41.00% in the number of parameters and the number of operations are respectively achieved, while increasing the test error only by less than 1%.

Trainable Spectrally Initializable Matrix Transformations in Convolutional Neural Networks

Michele Alberti, Angela Botros, Schuetz Narayan, Rolf Ingold, Marcus Liwicki, Mathias Seuret

Auto-TLDR; Trainable and Spectrally Initializable Matrix Transformations for Neural Networks

Abstract Slides Poster Similar

In this work, we introduce a new architectural component to Neural Networks (NN), i.e., trainable and spectrally initializable matrix transformations on feature maps. While previous literature has already demonstrated the possibility of adding static spectral transformations as feature processors, our focus is on more general trainable transforms. We study the transforms in various architectural configurations on four datasets of different nature: from medical (ColorectalHist, HAM10000) and natural (Flowers) images to historical documents (CB55). With rigorous experiments that control for the number of parameters and randomness, we show that networks utilizing the introduced matrix transformations outperform vanilla neural networks. The observed accuracy increases appreciably across all datasets. In addition, we show that the benefit of spectral initialization leads to significantly faster convergence, as opposed to randomly initialized matrix transformations. The transformations are implemented as auto-differentiable PyTorch modules that can be incorporated into any neural network architecture. The entire code base is open-source.

Attention As Activation

Yimian Dai, Stefan Oehmcke, Fabian Gieseke, Yiquan Wu, Kobus Barnard

Auto-TLDR; Attentional Activation Units for Convolutional Networks

Abstract Slides Similar

Activation functions and attention mechanisms are typically treated as having different purposes and have evolved differently. However, both concepts can be formulated as a non-linear gating function. Inspired by their similarity, we propose a novel type of activation units called attentional activation~(ATAC) units as a unification of activation functions and attention mechanisms. In particular, we propose a local channel attention module for the simultaneous non-linear activation and element-wise feature refinement, which locally aggregates point-wise cross-channel feature contexts. By replacing the well-known rectified linear units by such ATAC units in convolutional networks, we can construct fully attentional networks that perform significantly better with a modest number of additional parameters. We conducted detailed ablation studies on the ATAC units using several host networks with varying network depths to empirically verify the effectiveness and efficiency of the units. Furthermore, we compared the performance of the ATAC units against existing activation functions as well as other attention mechanisms on the CIFAR-10, CIFAR-100, and ImageNet datasets. Our experimental results show that networks constructed with the proposed ATAC units generally yield performance gains over their competitors given a comparable number of parameters.

Learning Sparse Deep Neural Networks Using Efficient Structured Projections on Convex Constraints for Green AI

Michel Barlaud, Frederic Guyard

Auto-TLDR; Constrained Deep Neural Network with Constrained Splitting Projection

Abstract Slides Poster Similar

In recent years, deep neural networks (DNN) have been applied to different domains and achieved dramatic performance improvements over state-of-the-art classical methods. These performances of DNNs were however often obtained with networks containing millions of parameters and which training required heavy computational power. In order to cope with this computational issue a huge literature deals with proximal regularization methods which are time consuming.\\ In this paper, we propose instead a constrained approach. We provide the general framework for our new splitting projection gradient method. Our splitting algorithm iterates a gradient step and a projection on convex sets. We study algorithms for different constraints: the classical $\ell_1$ unstructured constraint and structured constraints such as the nuclear norm, the $\ell_{2,1} $ constraint (Group LASSO). We propose a new $\ell_{1,1} $ structured constraint for which we provide a new projection algorithm We demonstrate the effectiveness of our method on three popular datasets (MNIST, Fashion MNIST and CIFAR). Experiments on these datasets show that our splitting projection method with our new $\ell_{1,1} $ structured constraint provides the best reduction of memory and computational power. Experiments show that fully connected linear DNN are more efficient for green AI.

Color, Edge, and Pixel-Wise Explanation of Predictions Based onInterpretable Neural Network Model

Jay Hoon Jung, Youngmin Kwon

Auto-TLDR; Explainable Deep Neural Network with Edge Detecting Filters

Abstract Poster Similar

We design an interpretable network model by introducing explainable components into a Deep Neural Network (DNN). We substituted the first kernels of a Convolutional Neural Network (CNN) and a ResNet-50 with the well-known edge detecting filters such as Sobel, Prewitt, and other filters. Each filters' relative importance scores are measured with a variant of Layer-wise Relevance Propagation (LRP) method proposed by Bach et al. Since the effects of the edge detecting filters are well understood, our model provides three different scores to explain individual predictions: the scores with respect to (1) colors, (2) edge filters, and (3) pixels of the image. Our method provides more tools to analyze the predictions by highlighting the location of important edges and colors in the images. Furthermore, the general features of a category can be shown in our scores as well as individual predictions. At the same time, the model does not degrade performances on MNIST, Fruit360 and ImageNet datasets.

Multi-Layered Discriminative Restricted Boltzmann Machine with Untrained Probabilistic Layer

Yuri Kanno, Muneki Yasuda

Auto-TLDR; MDRBM: A Probabilistic Four-layered Neural Network for Extreme Learning Machine

Abstract Poster Similar

An extreme learning machine (ELM) is a three-layered feed-forward neural network having untrained parameters, which are randomly determined before training. Inspired by the idea of ELM, a probabilistic untrained layer called a probabilistic-ELM (PELM) layer is proposed, and it is combined with a discriminative restricted Boltzmann machine (DRBM), which is a probabilistic three-layered neural network for solving classification problems. The proposed model is obtained by stacking DRBM on the PELM layer. The resultant model (i.e., multi-layered DRBM (MDRBM)) forms a probabilistic four-layered neural network. In MDRBM, the parameters in the PELM layer can be determined using Gaussian-Bernoulli restricted Boltzmann machine. Owing to the PELM layer, MDRBM obtains a strong immunity against noise in inputs, which is one of the most important advantages of MDRBM. Numerical experiments using some benchmark datasets, MNIST, Fashion-MNIST, Urban Land Cover, and CIFAR-10, demonstrate that MDRBM is superior to other existing models, particularly, in terms of the noise-robustness property (or, in other words, the generalization property).

A Multilinear Sampling Algorithm to Estimate Shapley Values

Ramin Okhrati, Aldo Lipani

Auto-TLDR; A sampling method for Shapley values for multilayer Perceptrons

Abstract Slides Poster Similar

Shapley values are great analytical tools in game theory to measure the importance of a player in a game. Due to their axiomatic and desirable properties such as efficiency, they have become popular for feature importance analysis in data science and machine learning. However, the time complexity to compute Shapley values based on the original formula is exponential, and as the number of features increases, this becomes infeasible. Castro et al. [1] developed a sampling algorithm, to estimate Shapley values. In this work, we propose a new sampling method based on a multilinear extension technique as applied in game theory. The aim is to provide a more efficient (sampling) method for estimating Shapley values. Our method is applicable to any machine learning model, in particular for either multiclass classifications or regression problems. We apply the method to estimate Shapley values for multilayer Perceptrons (MLPs) and through experimentation on two datasets, we demonstrate that our method provides more accurate estimations of the Shapley values by reducing the variance of the sampling statistics

Hcore-Init: Neural Network Initialization Based on Graph Degeneracy

Stratis Limnios, George Dasoulas, Dimitrios Thilikos, Michalis Vazirgiannis

Auto-TLDR; K-hypercore: Graph Mining for Deep Neural Networks

Abstract Slides Poster Similar

Neural networks are the pinnacle of Artificial Intelligence, as in recent years we witnessed many novel architectures, learning and optimization techniques for deep learning. Capitalizing on the fact that neural networks inherently constitute multipartite graphs among neuron layers, we aim to analyze directly their structure to extract meaningful information that can improve the learning process. To our knowledge graph mining techniques for enhancing learning in neural networks have not been thoroughly investigated. In this paper we propose an adapted version of the k-core structure for the complete weighted multipartite graph extracted from a deep learning architecture. As a multipartite graph is a combination of bipartite graphs, that are in turn the incidence graphs of hypergraphs, we design k-hypercore decomposition, the hypergraph analogue of k-core degeneracy. We applied k-hypercore to several neural network architectures, more specifically to convolutional neural networks and multilayer perceptrons for image recognition tasks after a very short pretraining. Then we used the information provided by the hypercore numbers of the neurons to re-initialize the weights of the neural network, thus biasing the gradient optimization scheme. Extensive experiments proved that k-hypercore outperforms the state-of-the-art initialization methods.

Improving Batch Normalization with Skewness Reduction for Deep Neural Networks

Pak Lun Kevin Ding, Martin Sarah, Baoxin Li

Auto-TLDR; Batch Normalization with Skewness Reduction

Abstract Slides Poster Similar

Batch Normalization (BN) is a well-known technique used in training deep neural networks. The main idea behind batch normalization is to normalize the features of the layers ($i.e.$, transforming them to have a mean equal to zero and a variance equal to one). Such a procedure encourages the optimization landscape of the loss function to be smoother, and improve the learning of the networks for both speed and performance. In this paper, we demonstrate that the performance of the network can be improved, if the distributions of the features of the output in the same layer are similar. As normalizing based on mean and variance does not necessarily make the features to have the same distribution, we propose a new normalization scheme: Batch Normalization with Skewness Reduction (BNSR). Comparing with other normalization approaches, BNSR transforms not just only the mean and variance, but also the skewness of the data. By tackling this property of a distribution, we are able to make the output distributions of the layers to be further similar. The nonlinearity of BNSR may further improve the expressiveness of the underlying network. Comparisons with other normalization schemes are tested on the CIFAR-100 and ImageNet datasets. Experimental results show that the proposed approach can outperform other state-of-the-arts that are not equipped with BNSR.

Improving Model Accuracy for Imbalanced Image Classification Tasks by Adding a Final Batch Normalization Layer: An Empirical Study

Veysel Kocaman, Ofer M. Shir, Thomas Baeck

Auto-TLDR; Exploiting Batch Normalization before the Output Layer in Deep Learning for Minority Class Detection in Imbalanced Data Sets

Abstract Slides Poster Similar

Some real-world domains, such as Agriculture and Healthcare, comprise early-stage disease indications whose recording constitutes a rare event, and yet, whose precise detection at that stage is critical. In this type of highly imbalanced classification problems, which encompass complex features, deep learning (DL) is much needed because of its strong detection capabilities. At the same time, DL is observed in practice to favor majority over minority classes and consequently suffer from inaccurate detection of the targeted early-stage indications. To simulate such scenarios, we artificially generate skewness (99% vs. 1%) for certain plant types out of the PlantVillage dataset as a basis for classification of scarce visual cues through transfer learning. By randomly and unevenly picking healthy and unhealthy samples from certain plant types to form a training set, we consider a base experiment as fine-tuning ResNet34 and VGG19 architectures and then testing the model performance on a balanced dataset of healthy and unhealthy images. We empirically observe that the initial F1 test score jumps from 0.29 to 0.95 for the minority class upon adding a final Batch Normalization (BN) layer just before the output layer in VGG19. We demonstrate that utilizing an additional BN layer before the output layer in modern CNN architectures has a considerable impact in terms of minimizing the training time and testing error for minority classes in highly imbalanced data sets. Moreover, when the final BN is employed, trying to minimize validation and training losses may not be an optimal way for getting a high F1 test score for minority classes in anomaly detection problems. That is, the network might perform better even if it is not ‘confident’ enough while making a prediction; leading to another discussion about why softmax output is not a good uncertainty measure for DL models.

Separation of Aleatoric and Epistemic Uncertainty in Deterministic Deep Neural Networks

Denis Huseljic, Bernhard Sick, Marek Herde, Daniel Kottke

Auto-TLDR; AE-DNN: Modeling Uncertainty in Deep Neural Networks

Abstract Slides Poster Similar

Despite the success of deep neural networks (DNN) in many applications, their ability to model uncertainty is still significantly limited. For example, in safety-critical applications such as autonomous driving, it is crucial to obtain a prediction that reflects different types of uncertainty to address life-threatening situations appropriately. In such cases, it is essential to be aware of the risk (i.e., aleatoric uncertainty) and the reliability (i.e., epistemic uncertainty) that comes with a prediction. We present AE-DNN, a model allowing the separation of aleatoric and epistemic uncertainty while maintaining a proper generalization capability. AE-DNN is based on deterministic DNN, which can determine the respective uncertainty measures in a single forward pass. In analyses with synthetic and image data, we show that our method improves the modeling of epistemic uncertainty while providing an intuitively understandable separation of risk and reliability.

Neuron-Based Network Pruning Based on Majority Voting

Ali Alqahtani, Xianghua Xie, Ehab Essa, Mark W. Jones

Auto-TLDR; Large-Scale Neural Network Pruning using Majority Voting

Abstract Slides Poster Similar

The achievement of neural networks in a variety of applications is accompanied by a dramatic increase in computational costs and memory requirements. In this paper, we propose an efficient method to simultaneously identify the critical neurons and prune the model during training without involving any pre-training or fine-tuning procedures. Unlike existing methods, which accomplish this task in a greedy fashion, we propose a majority voting technique to compare the activation values among neurons and assign a voting score to quantitatively evaluate their importance.This mechanism helps to effectively reduce model complexity by eliminating the less influential neurons and aims to determine a subset of the whole model that can represent the reference model with much fewer parameters within the training process. Experimental results show that majority voting efficiently compresses the network with no drop in model accuracy, pruning more than 79\% of the original model parameters on CIFAR10 and more than 91\% of the original parameters on MNIST. Moreover, we show that with our proposed method, sparse models can be further pruned into even smaller models by removing more than 60\% of the parameters, whilst preserving the reference model accuracy.

Efficient Online Subclass Knowledge Distillation for Image Classification

Maria Tzelepi, Nikolaos Passalis, Anastasios Tefas

Auto-TLDR; OSKD: Online Subclass Knowledge Distillation

Abstract Slides Poster Similar

Deploying state-of-the-art deep learning models on embedded systems dictates certain storage and computation limitations. During the recent few years Knowledge Distillation (KD) has been recognized as a prominent approach to address this issue. That is, KD has been effectively proposed for training fast and compact deep learning models by transferring knowledge from more complex and powerful models. However, knowledge distillation, in its conventional form, involves multiple stages of training, rendering it a computationally and memory demanding procedure. In this paper, a novel single-stage self knowledge distillation method is proposed, namely Online Subclass Knowledge Distillation (OSKD), that aims at revealing the similarities inside classes, improving the performance of any deep neural model in an online manner. Hence, as opposed to existing online distillation methods, we are able to acquire further knowledge from the model itself, without building multiple identical models or using multiple models to teach each other, rendering the OSKD approach more efficient. The experimental evaluation on two datasets validates that the proposed method improves the classification performance.

Neural Compression and Filtering for Edge-assisted Real-time Object Detection in Challenged Networks

Yoshitomo Matsubara, Marco Levorato

Auto-TLDR; Deep Neural Networks for Remote Object Detection Using Edge Computing

Abstract Slides Poster Similar

The edge computing paradigm places compute-capable devices - edge servers - at the network edge to assist mobile devices in executing data analysis tasks. Intuitively, offloading compute-intense tasks to edge servers can reduce their execution time. However, poor conditions of the wireless channel connecting the mobile devices to the edge servers may degrade the overall capture-to-output delay achieved by edge offloading. Herein, we focus on edge computing supporting remote object detection by means of Deep Neural Networks (DNN), and develop a framework to reduce the amount of data transmitted over the wireless link. The core idea we propose builds on recent approaches splitting DNNs into sections - namely head and tail models - executed by the mobile device and edge server, respectively. The wireless link, then, is used to transport the output of the last layer of the head model to the edge server, instead of the DNN input. Most prior work focuses on classification tasks and leaves the DNN structure unaltered. Herein, we focus on DNNs for three different object detection tasks, which present a much more convoluted structure, and modify the architecture of the network to: (i) achieve in-network compression by introducing a bottleneck layer in the early layers on the head model, and (ii) prefilter pictures that do not contain objects of interest using a convolutional neural network. Results show that the proposed technique represents an effective intermediate option between local and edge computing in a parameter region where these extreme point solutions fail to provide satisfactory performance.

HFP: Hardware-Aware Filter Pruning for Deep Convolutional Neural Networks Acceleration

Fang Yu, Chuanqi Han, Pengcheng Wang, Ruoran Huang, Xi Huang, Li Cui

Auto-TLDR; Hardware-Aware Filter Pruning for Convolutional Neural Networks

Abstract Slides Poster Similar

Convolutional Neural Networks (CNNs) are powerful but computationally demanding and memory intensive, thus impeding their practical applications on resource-constrained hardware. Filter pruning is an efficient approach for deep CNN compression and acceleration, which aims to eliminate some filters with tolerable performance degradation. In the literature, the majority of approaches prune networks by defining the redundant filters or training the networks with a sparsity prior loss function. These approaches mainly use FLOPs as their speed metric. However, the inference latency of pruned networks cannot be directly controlled on the hardware platform, which is an important dimension of practicality. To address this issue, we propose a novel Hardware-aware Filter Pruning method (HFP) which can produce pruned networks that satisfy the actual latency budget on the hardwares of interest. In addition, we propose an iterative pruning framework called Opti-Cut to decrease the accuracy degradation of pruning process and accelerate the pruning procedure whilst meeting the hardware budget. More specifically, HFP first builds up a lookup table for fast estimating the latency of target network about filter configuration layer by layer. Then, HFP leverages information gain (IG) to globally evaluate the filters' contribution to network output distribution. HFP utilizes the Opti-Cut framework to globally prune filters with the minimum IG one by one until the latency budget is satisfied. We verify the effectiveness of the proposed method on CIFAR-10 and ImageNet. Compared with the state-of-the-art pruning methods, HFP demonstrates superior performances on VGGNet, ResNet and MobileNet V1/V2.

Decision Snippet Features

Pascal Welke, Fouad Alkhoury, Christian Bauckhage, Stefan Wrobel

Auto-TLDR; Decision Snippet Features for Interpretability

Abstract Slides Poster Similar

Decision trees excel at interpretability of their prediction results. To achieve required prediction accuracies, however, often large ensembles of decision trees -- random forests -- are considered, reducing interpretability due to large size. Additionally, their size slows down inference on modern hardware and restricts their applicability in low-memory embedded devices. We introduce \emph{Decision Snippet Features}, which are obtained from small subtrees that appear frequently in trained random forests. We subsequently show that linear models on top of these features achieve comparable and sometimes even better predictive performance than the original random forest, while reducing the model size by up to two orders of magnitude.

Deep Transformation Models: Tackling Complex Regression Problems with Neural Network Based Transformation Models

Beate Sick, Torsten Hothorn, Oliver Dürr

Auto-TLDR; A Deep Transformation Model for Probabilistic Regression

Abstract Slides Poster Similar

We present a deep transformation model for probabilistic regression. Deep learning is known for outstandingly accurate predictions on complex data but in regression tasks it is predominantly used to just predict a single number. This ignores the non-deterministic character of most tasks. Especially if crucial decisions are based on the predictions, like in medical applications, it is essential to quantify the prediction uncertainty. The presented deep learning transformation model estimates the whole conditional probability distribution, which is the most thorough way to capture uncertainty about the outcome. We combine ideas from a statistical transformation model (most likely transformation) with recent transformation models from deep learning (normalizing flows) to predict complex outcome distributions. The core of the method is a parameterized transformation function which can be trained with the usual maximum likelihood framework using gradient descent. The method can be combined with existing deep learning architectures. For small machine learning benchmark datasets, we report state of the art performance for most dataset and partly even outperform it. Our method works for complex input data, which we demonstrate by employing a CNN architecture on image data.

Probability Guided Maxout

Claudio Ferrari, Stefano Berretti, Alberto Del Bimbo

Auto-TLDR; Probability Guided Maxout for CNN Training

Abstract Slides Poster Similar

In this paper, we propose an original CNN training strategy that brings together ideas from both dropout-like regularization methods and solutions that learn discriminative features. We propose a dropping criterion that, differently from dropout and its variants, is deterministic rather than random. It grounds on the empirical evidence that feature descriptors with larger $L2$-norm and highly-active nodes are strongly correlated to confident class predictions. Thus, our criterion guides towards dropping a percentage of the most active nodes of the descriptors, proportionally to the estimated class probability. We simultaneously train a per-sample scaling factor to balance the expected output across training and inference. This further allows us to keep high the descriptor's L2-norm, which we show enforces confident predictions. The combination of these two strategies resulted in our ``Probability Guided Maxout'' solution that acts as a training regularizer. We prove the above behaviors by reporting extensive image classification results on the CIFAR10, CIFAR100, and Caltech256 datasets.

Is the Meta-Learning Idea Able to Improve the Generalization of Deep Neural Networks on the Standard Supervised Learning?

Xiang Deng, Zhongfei Zhang

Auto-TLDR; Meta-learning Based Training of Deep Neural Networks for Few-Shot Learning

Abstract Slides Poster Similar

Substantial efforts have been made on improving the generalization abilities of deep neural networks (DNNs) in order to obtain better performances without introducing more parameters. On the other hand, meta-learning approaches exhibit powerful generalization on new tasks in few-shot learning. Intuitively, few-shot learning is more challenging than the standard supervised learning as each target class only has a very few or no training samples. The natural question that arises is whether the meta-learning idea can be used for improving the generalization of DNNs on the standard supervised learning. In this paper, we propose a novel meta-learning based training procedure (MLTP) for DNNs and demonstrate that the meta-learning idea can indeed improve the generalization abilities of DNNs. MLTP simulates the meta-training process by considering a batch of training samples as a task. The key idea is that the gradient descent step for improving the current task performance should also improve a new task performance, which is ignored by the current standard procedure for training neural networks. MLTP also benefits from all the existing training techniques such as dropout, weight decay, and batch normalization. We evaluate MLTP by training a variety of small and large neural networks on three benchmark datasets, i.e., CIFAR-10, CIFAR-100, and Tiny ImageNet. The experimental results show a consistently improved generalization performance on all the DNNs with different sizes, which verifies the promise of MLTP and demonstrates that the meta-learning idea is indeed able to improve the generalization of DNNs on the standard supervised learning.

VOWEL: A Local Online Learning Rule for Recurrent Networks of Probabilistic Spiking Winner-Take-All Circuits

Hyeryung Jang, Nicolas Skatchkovsky, Osvaldo Simeone

Auto-TLDR; VOWEL: A Variational Online Local Training Rule for Winner-Take-All Spiking Neural Networks

Abstract Slides Similar

Networks of spiking neurons and Winner-Take-All spiking circuits (WTA-SNNs) can detect information encoded in spatio-temporal multi-valued events. These are described by the timing of events of interest, e.g., clicks, as well as by categorical numerical values assigned to each event, e.g., like or dislike. Other use cases include object recognition from data collected by neuromorphic cameras, which produce, for each pixel, signed bits at the times of sufficiently large brightness variations. Existing schemes for training WTA-SNNs are limited to rate-encoding solutions, and are hence able to detect only spatial patterns. Developing more general training algorithms for arbitrary WTA-SNNs inherits the challenges of training (binary) Spiking Neural Networks (SNNs). These amount, most notably, to the non-differentiability of threshold functions, to the recurrent behavior of spiking neural models, and to the difficulty of implementing backpropagation in neuromorphic hardware. In this paper, we develop a variational online local training rule for WTA-SNNs, referred to as VOWEL, that leverages only local pre- and post-synaptic information for visible circuits, and an additional common reward signal for hidden circuits. The method is based on probabilistic generalized linear neural models, control variates, and variational regularization. Experimental results on real-world neuromorphic datasets with multi-valued events demonstrate the advantages of WTA-SNNs over conventional binary SNNs trained with state-of-the-art methods, especially in the presence of limited computing resources.

Feature-Dependent Cross-Connections in Multi-Path Neural Networks

Dumindu Tissera, Kasun Vithanage, Rukshan Wijesinghe, Kumara Kahatapitiya, Subha Fernando, Ranga Rodrigo

Auto-TLDR; Multi-path Networks for Adaptive Feature Extraction

Abstract Slides Poster Similar

Learning a particular task from a dataset, samples in which originate from diverse contexts, is challenging, and usually addressed by deepening or widening standard neural networks. As opposed to conventional network widening, multi-path architectures restrict the quadratic increment of complexity to a linear scale. However, existing multi-column/path networks or model ensembling methods do not consider any feature-dependant allocation of parallel resources, and therefore, tend to learn redundant features. Given a layer in a multi-path network, if we restrict each path to learn a context-specific set of features and introduce a mechanism to intelligently allocate incoming feature maps to such paths, each path can specialize in a certain context, reducing the redundancy and improving the quality of extracted features. This eventually leads to better-optimized usage of parallel resources. To do this, we propose inserting feature-dependant cross-connections between parallel sets of feature maps in successive layers. The weights of these cross-connections are learned based on the input features of the particular layer. Our multi-path networks show improved image recognition accuracy at a similar complexity compared to conventional and state-of-the-art methods for deepening, widening and adaptive feature extracting, in both small and large scale datasets.

An Efficient Empirical Solver for Localized Multiple Kernel Learning Via DNNs

Ziming Zhang

Auto-TLDR; Localized Multiple Kernel Learning using LMKL-Net

Abstract Slides Poster Similar

In this paper we propose solving localized multiple kernel learning (LMKL) using LMKL-Net, a feedforward deep neural network (DNN). In contrast to previous works, as a learning principle we propose parameterizing the gating function for learning kernel combination weights and the multiclass classifier using an attentional network (AN) and a multilayer perceptron (MLP), respectively. Such interpretability helps us better understand how the network solves the problem. Thanks to stochastic gradient descent (SGD), our approach has {\em linear} computational complexity in training. Empirically on benchmark datasets we demonstrate that with comparable or better accuracy than the state-of-the-art, our LMKL-Net can be trained about {\bf two orders of magnitude} faster with about {\bf two orders of magnitude} smaller memory footprint for large-scale learning.

Learning to Prune in Training via Dynamic Channel Propagation

Shibo Shen, Rongpeng Li, Zhifeng Zhao, Honggang Zhang, Yugeng Zhou

Auto-TLDR; Dynamic Channel Propagation for Neural Network Pruning

Abstract Slides Poster Similar

In this paper, we propose a novel network training mechanism called "dynamic channel propagation" to prune the model during the training period. In particular, we pick up a specific group of channels in each convolutional layer to participate in the forward propagation in training time according to the significance level of channel, which is defined as channel utility. The utility values with respect to all selected channels are updated simultaneously with the error back-propagation process and will constantly change. Furthermore, when the training ends, channels with high utility values are retained whereas those with low utility values are discarded. Hence, our proposed method trains and prunes neural networks simultaneously. We empirically evaluate our novel training method on various representative benchmark datasets and advanced convolutional neural network (CNN) architectures, including VGGNet and ResNet. The experiment results verify superior performance and robust effectiveness of our approach.

Verifying the Causes of Adversarial Examples

Honglin Li, Yifei Fan, Frieder Ganz, Tony Yezzi, Payam Barnaghi

Auto-TLDR; Exploring the Causes of Adversarial Examples in Neural Networks

Abstract Slides Poster Similar

The robustness of neural networks is challenged by adversarial examples that contain almost imperceptible perturbations to inputs which mislead a classifier to incorrect outputs in high confidence. Limited by the extreme difficulty in examining a high-dimensional image space thoroughly, research on explaining and justifying the causes of adversarial examples falls behind studies on attacks and defenses. In this paper, we present a collection of potential causes of adversarial examples and verify (or partially verify) them through carefully-designed controlled experiments. The major causes of adversarial examples include model linearity, one-sum constraint, and geometry of the categories. To control the effect of those causes, multiple techniques are applied such as $L_2$ normalization, replacement of loss functions, construction of reference datasets, and novel models using multi-layer perceptron probabilistic neural networks (MLP-PNN) and density estimation (DE). Our experiment results show that geometric factors tend to be more direct causes and statistical factors magnify the phenomenon, especially for assigning high prediction confidence. We hope this paper will inspire more studies to rigorously investigate the root causes of adversarial examples, which in turn provide useful guidance on designing more robust models.

Dynamic Multi-Path Neural Network

Yingcheng Su, Yichao Wu, Ken Chen, Ding Liang, Xiaolin Hu

Auto-TLDR; Dynamic Multi-path Neural Network

Abstract Slides Similar

Although deeper and larger neural networks have achieved better performance, due to overwhelming burden on computation, they cannot meet the demands of deployment on resource-limited devices. An effective strategy to address this problem is to make use of dynamic inference mechanism, which changes the inference path for different samples at runtime. Existing methods only reduce the depth by skipping an entire specific layer, which may lose important information in this layer. In this paper, we propose a novel method called Dynamic Multi-path Neural Network (DMNN), which provides more topology choices in terms of both width and depth on the fly. For better modelling the inference path selection, we further introduce previous state and object category information to guide the training process. Compared to previous dynamic inference techniques, the proposed method is more flexible and easier to incorporate into most modern network architectures. Experimental results on ImageNet and CIFAR-100 demonstrate the superiority of our method on both efficiency and classification accuracy.

Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks

Sebastian Palacio, Philipp Engler, Jörn Hees, Andreas Dengel

Auto-TLDR; Self-Supervised Autogenous Learning for Deep Neural Networks

Abstract Slides Poster Similar

Classification problems solved with deep neural networks (DNNs) typically rely on a closed world paradigm, and optimize over a single objective (e.g., minimization of the cross- entropy loss). This setup dismisses all kinds of supporting signals that can be used to reinforce the existence or absence of particular patterns. The increasing need for models that are interpretable by design makes the inclusion of said contextual signals a crucial necessity. To this end, we introduce the notion of Self-Supervised Autogenous Learning (SSAL). A SSAL objective is realized through one or more additional targets that are derived from the original supervised classification task, following architectural principles found in multi-task learning. SSAL branches impose low-level priors into the optimization process (e.g., grouping). The ability of using SSAL branches during inference, allow models to converge faster, focusing on a richer set of class-relevant features. We equip state-of-the-art DNNs with SSAL objectives and report consistent improvements for all of them on CIFAR100 and Imagenet. We show that SSAL models outperform similar state-of-the-art methods focused on contextual loss functions, auxiliary branches and hierarchical priors.

RNN Training along Locally Optimal Trajectories via Frank-Wolfe Algorithm

Yun Yue, Ming Li, Venkatesh Saligrama, Ziming Zhang

Auto-TLDR; Frank-Wolfe Algorithm for Efficient Training of RNNs

Abstract Slides Poster Similar

We propose a novel and efficient training method for RNNs by iteratively seeking a local minima on the loss surface within a small region, and leverage this directional vector for the update, in an outer-loop. We propose to utilize the Frank-Wolfe (FW) algorithm in this context. Although, FW implicitly involves normalized gradients, which can lead to a slow convergence rate, we develop a novel RNN training method that, surprisingly, even with the additional cost, the overall training cost is empirically observed to be lower than back-propagation. Our method leads to a new Frank-Wolfe method, that is in essence an SGD algorithm with a restart scheme. We prove that under certain conditions our algorithm has a sublinear convergence rate of $O(1/\epsilon)$ for $\epsilon$ error. We then conduct empirical experiments on several benchmark datasets including those that exhibit long-term dependencies, and show significant performance improvement. We also experiment with deep RNN architectures and show efficient training performance. Finally, we demonstrate that our training method is robust to noisy data.

Exploiting Non-Linear Redundancy for Neural Model Compression

Muhammad Ahmed Shah, Raphael Olivier, Bhiksha Raj

Auto-TLDR; Compressing Deep Neural Networks with Linear Dependency

Abstract Slides Poster Similar

Deploying deep learning models with millions, even billions, of parameters is challenging given real world memory, power and compute constraints. In an effort to make these models more practical, in this paper, we propose a novel model compression approach that exploits linear dependence between the activations in a layer to eliminate entire structural units (neurons/convolutional filters). Our approach also adjusts the weights of the layer in a manner that is provably lossless while training if the removed neuron was perfectly predictable. We combine this approach with an annealing algorithm that may be applied during training, or even on a trained model, and demonstrate, using popular datasets, that our technique can reduce the parameters of VGG and AlexNet by more than 97\% on \cifar, 85\% on \caltech, and 19\% on ImageNet at less than 2\% loss in accuracy. Furthermore, we provide theoretical results showing that in overparametrized, locally linear (ReLU) neural networks where redundant features exist, and with correct hyperparameter selection, our method is indeed able to capture and suppress those dependencies.

WeightAlign: Normalizing Activations by Weight Alignment

Xiangwei Shi, Yunqiang Li, Xin Liu, Jan Van Gemert

Auto-TLDR; WeightAlign: Normalization of Activations without Sample Statistics

Abstract Slides Poster Similar

Batch normalization (BN) allows training very deep networks by normalizing activations by mini-batch sample statistics which renders BN unstable for small batch sizes. Current small-batch solutions such as Instance Norm, Layer Norm, and Group Norm use channel statistics which can be computed even for a single sample. Such methods are less stable than BN as they critically depend on the statistics of a single input sample. To address this problem, we propose a normalization of activation without sample statistics. We present WeightAlign: a method that normalizes the weights by the mean and scaled standard derivation computed within a filter, which normalizes activations without computing any sample statistics. Our proposed method is independent of batch size and stable over a wide range of batch sizes. Because weight statistics are orthogonal to sample statistics, we can directly combine WeightAlign with any method for activation normalization. We experimentally demonstrate these benefits for classification on CIFAR-10, CIFAR-100, ImageNet, for semantic segmentation on PASCAL VOC 2012 and for domain adaptation on Office-31.

Low-Cost Lipschitz-Independent Adaptive Importance Sampling of Stochastic Gradients

Huikang Liu, Xiaolu Wang, Jiajin Li, Man-Cho Anthony So

Auto-TLDR; Adaptive Importance Sampling for Stochastic Gradient Descent

Abstract Slides Similar

Stochastic gradient descent (SGD) usually samples training data based on the uniform distribution, which may not be a good choice because of the high variance of its stochastic gradient. Thus, importance sampling methods are considered in the literature to improve the performance. Most previous work on SGD-based methods with importance sampling requires the knowledge of Lipschitz constants of all component gradients, which are in general difficult to estimate. In this paper, we study an adaptive importance sampling method for common SGD-based methods by exploiting the local first-order information without knowing any Lipschitz constants. In particular, we periodically changes the sampling distribution by only utilizing the gradient norms in the past few iterations. We prove that our adaptive importance sampling non-asymptotically reduces the variance of the stochastic gradients in SGD, and thus better convergence bounds than that for vanilla SGD can be obtained. We extend this sampling method to several other widely used stochastic gradient algorithms including SGD with momentum and ADAM. Experiments on common convex learning problems and deep neural networks illustrate notably enhanced performance using the adaptive sampling strategy.

Norm Loss: An Efficient yet Effective Regularization Method for Deep Neural Networks

Theodoros Georgiou, Sebastian Schmitt, Thomas Baeck, Wei Chen, Michael Lew

Auto-TLDR; Weight Soft-Regularization with Oblique Manifold for Convolutional Neural Network Training

Abstract Slides Poster Similar

Convolutional neural network training can suffer from diverse issues like exploding or vanishing gradients, scaling-based weight space symmetry and covariant-shift. In order to address these issues, researchers develop weight regularization methods and activation normalization methods. In this work we propose a weight soft-regularization method based on the Oblique manifold. The proposed method uses a loss function which pushes each weight vector to have a norm close to one, i.e. the weight matrix is smoothly steered toward the so-called Oblique manifold. We evaluate our method on the very popular CIFAR-10, CIFAR-100 and ImageNet 2012 datasets using two state-of-the-art architectures, namely the ResNet and wide-ResNet. Our method introduces negligible computational overhead and the results show that it is competitive to the state-of-the-art and in some cases superior to it. Additionally, the results are less sensitive to hyperparameter settings such as batch size and regularization factor.

Compact CNN Structure Learning by Knowledge Distillation

Waqar Ahmed, Andrea Zunino, Pietro Morerio, Vittorio Murino

Auto-TLDR; Knowledge Distillation for Compressing Deep Convolutional Neural Networks

Abstract Slides Poster Similar

The concept of compressing deep Convolutional Neural Networks (CNNs) is essential to use limited computation, power, and memory resources on embedded devices. However, existing methods achieve this objective at the cost of a drop in inference accuracy in computer vision tasks. To address such a drawback, we propose a framework that leverages knowledge distillation along with customizable block-wise optimization to learn a lightweight CNN structure while preserving better control over the compression-performance tradeoff. Considering specific resource constraints, e.g., floating-point operations per second (FLOPs) or model-parameters, our method results in a state of the art network compression while being capable of achieving better inference accuracy. In a comprehensive evaluation, we demonstrate that our method is effective, robust, and consistent with results over a variety of network architectures and datasets, at negligible training overhead. In particular, for the already compact network MobileNet_v2, our method offers up to 2x and 5.2x better model compression in terms of FLOPs and model-parameters, respectively, while getting 1.05% better model performance than the baseline network.

Regularized Flexible Activation Function Combinations for Deep Neural Networks

Renlong Jie, Junbin Gao, Andrey Vasnev, Minh-Ngoc Tran

Auto-TLDR; Flexible Activation in Deep Neural Networks using ReLU and ELUs

Abstract Slides Poster Similar

Activation in deep neural networks is fundamental to achieving non-linear mappings. Traditional studies mainly focus on ﬁnding ﬁxed activations for a particular set of learning tasks or model architectures. The research on ﬂexible activation is quite limited in both designing philosophy and application scenarios. In this study, three principles of choosing ﬂexible activation components are proposed and a general combined form of ﬂexible activation functions is implemented. Based on this, a novel family of ﬂexible activation functions that can replace sigmoid or tanh in LSTM cells are implemented, as well as a new family by combining ReLU and ELUs. Also, two new regularisation terms based on assumptions as prior knowledge are introduced. It has been shown that LSTM models with proposed ﬂexible activations P-Sig-Ramp provide signiﬁcant improvements in time series forecasting, while the proposed P-E2-ReLU achieves better and more stable performance on lossy image compression tasks with convolutional auto-encoders. In addition, the proposed regularization terms improve the convergence,performance and stability of the models with ﬂexible activation functions. The code for this paper is available at https://github.com/9NXJRDDRQK/Flexible Activation.

Automatically Mining Relevant Variable Interactions Via Sparse Bayesian Learning

Ryoichiro Yafune, Daisuke Sakuma, Yasuo Tabei, Noritaka Saito, Hiroto Saigo

Auto-TLDR; Sparse Bayes for Interpretable Non-linear Prediction

Abstract Slides Poster Similar

With the rapid increase in the availability of large amount of data, prediction is becoming increasingly popular, and has widespread through our daily life. However, powerful non- linear prediction methods such as deep learning and SVM suffer from interpretability problem, making it hard to use in domains where the reason for decision making is required. In this paper, we develop an interpretable non-linear model called itemset Sparse Bayes (iSB), which builds a Bayesian probabilistic model, while simultaneously considering variable interactions. In order to suppress the resulting large number of variables, sparsity is imposed on regression weights by a sparsity inducing prior. As a subroutine to search for variable interactions, itemset enumeration algorithm is employed with a novel bounding condition. In computational experiments using real-world dataset, the proposed method performed better than decision tree by 10% in terms of r-squared . We also demonstrated the advantage of our method in Bayesian optimization setting, in which the proposed approach could successfully find the maximum of an unknown function faster than Gaussian process. The interpretability of iSB is naturally inherited to Bayesian optimization, thereby gives us a clue to understand which variables interactions are important in optimizing an unknown function.

Attention Based Pruning for Shift Networks

Ghouthi Hacene, Carlos Lassance, Vincent Gripon, Matthieu Courbariaux, Yoshua Bengio

Auto-TLDR; Shift Attention Layers for Efficient Convolutional Layers

Abstract Slides Poster Similar

In many application domains such as computer vision, Convolutional Layers (CLs) are key to the accuracy of deep learning methods. However, it is often required to assemble a large number of CLs, each containing thousands of parameters, in order to reach state-of-the-art accuracy, thus resulting in complex and demanding systems that are poorly fitted to resource-limited devices. Recently, methods have been proposed to replace the generic convolution operator by the combination of a shift operation and a simpler 1x1 convolution. The resulting block, called Shift Layer (SL), is an efficient alternative to CLs in the sense it allows to reach similar accuracies on various tasks with faster computations and fewer parameters. In this contribution, we introduce Shift Attention Layers (SALs), which extend SLs by using an attention mechanism that learns which shifts are the best at the same time the network function is trained. We demonstrate SALs are able to outperform vanilla SLs (and CLs) on various object recognition benchmarks while significantly reducing the number of float operations and parameters for the inference.