Paper Group ANR 140
A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks. Toward Understanding Catastrophic Forgetting in Continual Learning. Weight Friction: A Simple Method to Overcome Catastrophic Forgetting and Enable Continual Learning. TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications. Meta …
A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks
| Title | A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks |
| Authors | Saima Sharmin, Priyadarshini Panda, Syed Shakib Sarwar, Chankyu Lee, Wachirawit Ponghiran, Kaushik Roy |
| Abstract | In this era of machine learning models, their functionality is being threatened by adversarial attacks. In the face of this struggle for making artificial neural networks robust, finding a model, resilient to these attacks, is very important. In this work, we present, for the first time, a comprehensive analysis of the behavior of more bio-plausible networks, namely Spiking Neural Network (SNN) under state-of-the-art adversarial tests. We perform a comparative study of the accuracy degradation between conventional VGG-9 Artificial Neural Network (ANN) and equivalent spiking network with CIFAR-10 dataset in both whitebox and blackbox setting for different types of single-step and multi-step FGSM (Fast Gradient Sign Method) attacks. We demonstrate that SNNs tend to show more resiliency compared to ANN under black-box attack scenario. Additionally, we find that SNN robustness is largely dependent on the corresponding training mechanism. We observe that SNNs trained by spike-based backpropagation are more adversarially robust than the ones obtained by ANN-to-SNN conversion rules in several whitebox and blackbox scenarios. Finally, we also propose a simple, yet, effective framework for crafting adversarial attacks from SNNs. Our results suggest that attacks crafted from SNNs following our proposed method are much stronger than those crafted from ANNs. |
| Tasks | |
| Published | 2019-05-07 |
| URL | https://arxiv.org/abs/1905.02704v1 |
| https://arxiv.org/pdf/1905.02704v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-comprehensive-analysis-on-adversarial |
| Repo | |
| Framework | |
Toward Understanding Catastrophic Forgetting in Continual Learning
| Title | Toward Understanding Catastrophic Forgetting in Continual Learning |
| Authors | Cuong V. Nguyen, Alessandro Achille, Michael Lam, Tal Hassner, Vijay Mahadevan, Stefano Soatto |
| Abstract | We study the relationship between catastrophic forgetting and properties of task sequences. In particular, given a sequence of tasks, we would like to understand which properties of this sequence influence the error rates of continual learning algorithms trained on the sequence. To this end, we propose a new procedure that makes use of recent developments in task space modeling as well as correlation analysis to specify and analyze the properties we are interested in. As an application, we apply our procedure to study two properties of a task sequence: (1) total complexity and (2) sequential heterogeneity. We show that error rates are strongly and positively correlated to a task sequence’s total complexity for some state-of-the-art algorithms. We also show that, surprisingly, the error rates have no or even negative correlations in some cases to sequential heterogeneity. Our findings suggest directions for improving continual learning benchmarks and methods. |
| Tasks | Continual Learning |
| Published | 2019-08-02 |
| URL | https://arxiv.org/abs/1908.01091v1 |
| https://arxiv.org/pdf/1908.01091v1.pdf | |
| PWC | https://paperswithcode.com/paper/toward-understanding-catastrophic-forgetting |
| Repo | |
| Framework | |
Weight Friction: A Simple Method to Overcome Catastrophic Forgetting and Enable Continual Learning
| Title | Weight Friction: A Simple Method to Overcome Catastrophic Forgetting and Enable Continual Learning |
| Authors | Gabrielle K. Liu |
| Abstract | In recent years, deep neural networks have found success in replicating human-level cognitive skills, yet they suffer from several major obstacles. One significant limitation is the inability to learn new tasks without forgetting previously learned tasks, a shortcoming known as catastrophic forgetting. In this research, we propose a simple method to overcome catastrophic forgetting and enable continual learning in neural networks. We draw inspiration from principles in neurology and physics to develop the concept of weight friction. Weight friction operates by a modification to the update rule in the gradient descent optimization method. It converges at a rate comparable to that of the stochastic gradient descent algorithm and can operate over multiple task domains. It performs comparably to current methods while offering improvements in computation and memory efficiency. |
| Tasks | Continual Learning |
| Published | 2019-08-02 |
| URL | https://arxiv.org/abs/1908.01052v2 |
| https://arxiv.org/pdf/1908.01052v2.pdf | |
| PWC | https://paperswithcode.com/paper/weight-friction-a-simple-method-to-overcome |
| Repo | |
| Framework | |
TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications
| Title | TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications |
| Authors | Yu-Chuan Huang, I-No Liao, Ching-Hsuan Chen, Tsì-Uí İk, Wen-Chih Peng |
| Abstract | Ball trajectory data are one of the most fundamental and useful information in the evaluation of players’ performance and analysis of game strategies. Although vision-based object tracking techniques have been developed to analyze sport competition videos, it is still challenging to recognize and position a high-speed and tiny ball accurately. In this paper, we develop a deep learning network, called TrackNet, to track the tennis ball from broadcast videos in which the ball images are small, blurry, and sometimes with afterimage tracks or even invisible. The proposed heatmap-based deep learning network is trained to not only recognize the ball image from a single frame but also learn flying patterns from consecutive frames. TrackNet takes images with a size of $640\times360$ to generate a detection heatmap from either a single frame or several consecutive frames to position the ball and can achieve high precision even on public domain videos. The network is evaluated on the video of the men’s singles final at the 2017 Summer Universiade, which is available on YouTube. The precision, recall, and F1-measure of TrackNet reach $99.7%$, $97.3%$, and $98.5%$, respectively. To prevent overfitting, 9 additional videos are partially labeled together with a subset from the previous dataset to implement 10-fold cross-validation, and the precision, recall, and F1-measure are $95.3%$, $75.7%$, and $84.3%$, respectively. A conventional image processing algorithm is also implemented to compare with TrackNet. Our experiments indicate that TrackNet outperforms conventional method by a big margin and achieves exceptional ball tracking performance. The dataset and demo video are available at https://nol.cs.nctu.edu.tw/ndo3je6av9/. |
| Tasks | Object Tracking |
| Published | 2019-07-08 |
| URL | https://arxiv.org/abs/1907.03698v1 |
| https://arxiv.org/pdf/1907.03698v1.pdf | |
| PWC | https://paperswithcode.com/paper/tracknet-a-deep-learning-network-for-tracking |
| Repo | |
| Framework | |
MetaMT,a MetaLearning Method Leveraging Multiple Domain Data for Low Resource Machine Translation
| Title | MetaMT,a MetaLearning Method Leveraging Multiple Domain Data for Low Resource Machine Translation |
| Authors | Rumeng Li, Xun Wang, Hong Yu |
| Abstract | Manipulating training data leads to robust neural models for MT. |
| Tasks | Machine Translation |
| Published | 2019-12-11 |
| URL | https://arxiv.org/abs/1912.05467v1 |
| https://arxiv.org/pdf/1912.05467v1.pdf | |
| PWC | https://paperswithcode.com/paper/metamta-metalearning-method-leveraging |
| Repo | |
| Framework | |
Example-Guided Style Consistent Image Synthesis from Semantic Labeling
| Title | Example-Guided Style Consistent Image Synthesis from Semantic Labeling |
| Authors | Miao Wang, Guo-Ye Yang, Ruilong Li, Run-Ze Liang, Song-Hai Zhang, Peter. M. Hall, Shi-Min Hu |
| Abstract | Example-guided image synthesis aims to synthesize an image from a semantic label map and an exemplary image indicating style. We use the term “style” in this problem to refer to implicit characteristics of images, for example: in portraits “style” includes gender, racial identity, age, hairstyle; in full body pictures it includes clothing; in street scenes, it refers to weather and time of day and such like. A semantic label map in these cases indicates facial expression, full body pose, or scene segmentation. We propose a solution to the example-guided image synthesis problem using conditional generative adversarial networks with style consistency. Our key contributions are (i) a novel style consistency discriminator to determine whether a pair of images are consistent in style; (ii) an adaptive semantic consistency loss; and (iii) a training data sampling strategy, for synthesizing style-consistent results to the exemplar. |
| Tasks | Image Generation, Scene Segmentation |
| Published | 2019-06-04 |
| URL | https://arxiv.org/abs/1906.01314v2 |
| https://arxiv.org/pdf/1906.01314v2.pdf | |
| PWC | https://paperswithcode.com/paper/example-guided-style-consistent-image-1 |
| Repo | |
| Framework | |
End-to-End Learning of Representations for Asynchronous Event-Based Data
| Title | End-to-End Learning of Representations for Asynchronous Event-Based Data |
| Authors | Daniel Gehrig, Antonio Loquercio, Konstantinos G. Derpanis, Davide Scaramuzza |
| Abstract | Event cameras are vision sensors that record asynchronous streams of per-pixel brightness changes, referred to as “events”. They have appealing advantages over frame-based cameras for computer vision, including high temporal resolution, high dynamic range, and no motion blur. Due to the sparse, non-uniform spatiotemporal layout of the event signal, pattern recognition algorithms typically aggregate events into a grid-based representation and subsequently process it by a standard vision pipeline, e.g., Convolutional Neural Network (CNN). In this work, we introduce a general framework to convert event streams into grid-based representations through a sequence of differentiable operations. Our framework comes with two main advantages: (i) allows learning the input event representation together with the task dedicated network in an end to end manner, and (ii) lays out a taxonomy that unifies the majority of extant event representations in the literature and identifies novel ones. Empirically, we show that our approach to learning the event representation end-to-end yields an improvement of approximately 12% on optical flow estimation and object recognition over state-of-the-art methods. |
| Tasks | Object Recognition, Optical Flow Estimation |
| Published | 2019-04-17 |
| URL | https://arxiv.org/abs/1904.08245v4 |
| https://arxiv.org/pdf/1904.08245v4.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-learning-of-representations-for |
| Repo | |
| Framework | |
Self-supervised pre-training with acoustic configurations for replay spoofing detection
| Title | Self-supervised pre-training with acoustic configurations for replay spoofing detection |
| Authors | Hye-jin Shim, Hee-Soo Heo, Jee-weon Jung, Ha-Jin Yu |
| Abstract | Large datasets are well-known as a key to the recent advances in deep learning. However, dataset construction, especially for replay spoofing detection, requires the physical process of playing an utterance and re-recording it, which hinders the construction of large-scale datasets. To compensate for the limited availability of replay spoofing datasets, in this study, we propose a method for pre-training acoustic configurations using external data unrelated to replay attacks. Here, acoustic configurations refer to variables present in the process of a voice being uttered by a speaker and recorded through a microphone. Specifically, we select pairs of audio segments and train the network to determine whether the acoustic configurations of two segments are identical. We conducted experiments using the ASVspoof 2019 physical access dataset, and the results revealed that our proposed method reduced the relative error rate by over 37% compared to the baseline. |
| Tasks | |
| Published | 2019-10-22 |
| URL | https://arxiv.org/abs/1910.09778v1 |
| https://arxiv.org/pdf/1910.09778v1.pdf | |
| PWC | https://paperswithcode.com/paper/self-supervised-pre-training-with-acoustic |
| Repo | |
| Framework | |
Multi-Frequency Vector Diffusion Maps
| Title | Multi-Frequency Vector Diffusion Maps |
| Authors | Yifeng Fan, Zhizhen Zhao |
| Abstract | We introduce multi-frequency vector diffusion maps (MFVDM), a new framework for organizing and analyzing high dimensional datasets. The new method is a mathematical and algorithmic generalization of vector diffusion maps (VDM) and other non-linear dimensionality reduction methods. MFVDM combines different nonlinear embeddings of the data points defined with multiple unitary irreducible representations of the alignment group that connect two nodes in the graph. We illustrate the efficacy of MFVDM on synthetic data generated according to a random graph model and cryo-electron microscopy image dataset. The new method achieves better nearest neighbor search and alignment estimation than the state-of-the-arts VDM and diffusion maps (DM) on extremely noisy data. |
| Tasks | Dimensionality Reduction |
| Published | 2019-06-06 |
| URL | https://arxiv.org/abs/1906.02605v1 |
| https://arxiv.org/pdf/1906.02605v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-frequency-vector-diffusion-maps |
| Repo | |
| Framework | |
An explanation method for Siamese neural networks
| Title | An explanation method for Siamese neural networks |
| Authors | Lev V. Utkin, Maxim S. Kovalev, Ernest M. Kasimov |
| Abstract | A new method for explaining the Siamese neural network is proposed. It uses the following main ideas. First, the explained feature vector is compared with the prototype of the corresponding class computed at the embedding level (the Siamese neural network output). The important features at this level are determined as features which are close to the same features of the prototype. Second, an autoencoder is trained in a special way in order to take into account the embedding level of the Si-amese network, and its decoder part is used for reconstructing input data with the corresponding changes. Numerical experiments with the well-known dataset MNIST illustrate the propose method. |
| Tasks | |
| Published | 2019-11-18 |
| URL | https://arxiv.org/abs/1911.07702v1 |
| https://arxiv.org/pdf/1911.07702v1.pdf | |
| PWC | https://paperswithcode.com/paper/an-explanation-method-for-siamese-neural |
| Repo | |
| Framework | |
Most Important Fundamental Rule of Poker Strategy
| Title | Most Important Fundamental Rule of Poker Strategy |
| Authors | Sam Ganzfried, Max Chiswick |
| Abstract | Poker is a large complex game of imperfect information, which has been singled out as a major AI challenge problem. Recently there has been a series of breakthroughs culminating in agents that have successfully defeated the strongest human players in two-player no-limit Texas hold ‘em. The strongest agents are based on algorithms for approximating Nash equilibrium strategies, which are stored in massive binary files and unintelligible to humans. A recent line of research has explored approaches for extrapolating knowledge from strong game-theoretic strategies that can be understood by humans. This would be useful when humans are the ultimate decision maker and allow humans to make better decisions from massive algorithmically-generated strategies. Using techniques from machine learning we have uncovered a new simple, fundamental rule of poker strategy that leads to a significant improvement in performance over the best prior rule and can also easily be applied by human players. |
| Tasks | Game of Poker |
| Published | 2019-06-08 |
| URL | https://arxiv.org/abs/1906.09895v3 |
| https://arxiv.org/pdf/1906.09895v3.pdf | |
| PWC | https://paperswithcode.com/paper/most-important-fundamental-rule-of-poker |
| Repo | |
| Framework | |
On the Inductive Bias of Neural Tangent Kernels
| Title | On the Inductive Bias of Neural Tangent Kernels |
| Authors | Alberto Bietti, Julien Mairal |
| Abstract | State-of-the-art neural networks are heavily over-parameterized, making the optimization algorithm a crucial ingredient for learning predictive models with good generalization properties. A recent line of work has shown that in a certain over-parameterized regime, the learning dynamics of gradient descent are governed by a certain kernel obtained at initialization, called the neural tangent kernel. We study the inductive bias of learning in such a regime by analyzing this kernel and the corresponding function space (RKHS). In particular, we study smoothness, approximation, and stability properties of functions with finite norm, including stability to image deformations in the case of convolutional networks, and compare to other known kernels for similar architectures. |
| Tasks | |
| Published | 2019-05-29 |
| URL | https://arxiv.org/abs/1905.12173v2 |
| https://arxiv.org/pdf/1905.12173v2.pdf | |
| PWC | https://paperswithcode.com/paper/on-the-inductive-bias-of-neural-tangent |
| Repo | |
| Framework | |
Adversarially Trained Autoencoders for Parallel-Data-Free Voice Conversion
| Title | Adversarially Trained Autoencoders for Parallel-Data-Free Voice Conversion |
| Authors | Orhan Ocal, Oguz H. Elibol, Gokce Keskin, Cory Stephenson, Anil Thomas, Kannan Ramchandran |
| Abstract | We present a method for converting the voices between a set of speakers. Our method is based on training multiple autoencoder paths, where there is a single speaker-independent encoder and multiple speaker-dependent decoders. The autoencoders are trained with an addition of an adversarial loss which is provided by an auxiliary classifier in order to guide the output of the encoder to be speaker independent. The training of the model is unsupervised in the sense that it does not require collecting the same utterances from the speakers nor does it require time aligning over phonemes. Due to the use of a single encoder, our method can generalize to converting the voice of out-of-training speakers to speakers in the training dataset. We present subjective tests corroborating the performance of our method. |
| Tasks | Voice Conversion |
| Published | 2019-05-09 |
| URL | https://arxiv.org/abs/1905.03864v1 |
| https://arxiv.org/pdf/1905.03864v1.pdf | |
| PWC | https://paperswithcode.com/paper/adversarially-trained-autoencoders-for |
| Repo | |
| Framework | |
When Machine Learning Meets Wireless Cellular Networks: Deployment, Challenges, and Applications
| Title | When Machine Learning Meets Wireless Cellular Networks: Deployment, Challenges, and Applications |
| Authors | Ursula Challita, Henrik A. Ryden, Hugo Tullberg |
| Abstract | Artificial intelligence (AI) powered wireless networks promise to revolutionize the conventional operation and structure of current networks from network design to infrastructure management, cost reduction, and user performance improvement. Empowering future networks with AI functionalities will enable a shift from reactive/incident driven operations to proactive/data driven operations. This paper provides an overview on the integration of AI functionalities in 5G and beyond networks. Key factors for successful AI integration such as data, real-time network intelligence, security, and augmentation of human intelligence are highlighted. We also summarize the various types of network intelligence as well as machine learning based air interface in future networks. Use case examples for the application of AI to the wireless domain are then summarized. We highlight on applications to the physical layer, radio resource management, mobility management, wireless security, and localization. |
| Tasks | |
| Published | 2019-11-08 |
| URL | https://arxiv.org/abs/1911.03585v1 |
| https://arxiv.org/pdf/1911.03585v1.pdf | |
| PWC | https://paperswithcode.com/paper/when-machine-learning-meets-wireless-cellular |
| Repo | |
| Framework | |
Estimating Skin Tone and Effects on Classification Performance in Dermatology Datasets
| Title | Estimating Skin Tone and Effects on Classification Performance in Dermatology Datasets |
| Authors | Newton M. Kinyanjui, Timothy Odonga, Celia Cintas, Noel C. F. Codella, Rameswar Panda, Prasanna Sattigeri, Kush R. Varshney |
| Abstract | Recent advances in computer vision and deep learning have led to breakthroughs in the development of automated skin image analysis. In particular, skin cancer classification models have achieved performance higher than trained expert dermatologists. However, no attempt has been made to evaluate the consistency in performance of machine learning models across populations with varying skin tones. In this paper, we present an approach to estimate skin tone in benchmark skin disease datasets, and investigate whether model performance is dependent on this measure. Specifically, we use individual typology angle (ITA) to approximate skin tone in dermatology datasets. We look at the distribution of ITA values to better understand skin color representation in two benchmark datasets: 1) the ISIC 2018 Challenge dataset, a collection of dermoscopic images of skin lesions for the detection of skin cancer, and 2) the SD-198 dataset, a collection of clinical images capturing a wide variety of skin diseases. To estimate ITA, we first develop segmentation models to isolate non-diseased areas of skin. We find that the majority of the data in the the two datasets have ITA values between 34.5{\deg} and 48{\deg}, which are associated with lighter skin, and is consistent with under-representation of darker skinned populations in these datasets. We also find no measurable correlation between performance of machine learning model and ITA values, though more comprehensive data is needed for further validation. |
| Tasks | Skin Cancer Classification |
| Published | 2019-10-29 |
| URL | https://arxiv.org/abs/1910.13268v1 |
| https://arxiv.org/pdf/1910.13268v1.pdf | |
| PWC | https://paperswithcode.com/paper/191013268 |
| Repo | |
| Framework | |
