Paper Group AWR 341
Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure. Region Mutual Information Loss for Semantic Segmentation. GazeCorrection:Self-Guided Eye Manipulation in the wild using Self-Supervised Generative Adversarial Networks. Probing Neural Network Comprehension of Natural Language Arguments. Reparameterizable Subset Sampling …
Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure
| Title | Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure |
| Authors | Fuli Feng, Xiangnan He, Jie Tang, Tat-Seng Chua |
| Abstract | Recent efforts show that neural networks are vulnerable to small but intentional perturbations on input features in visual classification tasks. Due to the additional consideration of connections between examples (\eg articles with citation link tend to be in the same class), graph neural networks could be more sensitive to the perturbations, since the perturbations from connected examples exacerbate the impact on a target example. Adversarial Training (AT), a dynamic regularization technique, can resist the worst-case perturbations on input features and is a promising choice to improve model robustness and generalization. However, existing AT methods focus on standard classification, being less effective when training models on graph since it does not model the impact from connected examples. In this work, we explore adversarial training on graph, aiming to improve the robustness and generalization of models learned on graph. We propose Graph Adversarial Training (GraphAT), which takes the impact from connected examples into account when learning to construct and resist perturbations. We give a general formulation of GraphAT, which can be seen as a dynamic regularization scheme based on the graph structure. To demonstrate the utility of GraphAT, we employ it on a state-of-the-art graph neural network model — Graph Convolutional Network (GCN). We conduct experiments on two citation graphs (Citeseer and Cora) and a knowledge graph (NELL), verifying the effectiveness of GraphAT which outperforms normal training on GCN by 4.51% in node classification accuracy. Codes are available via: https://github.com/fulifeng/GraphAT. |
| Tasks | Node Classification |
| Published | 2019-02-20 |
| URL | https://arxiv.org/abs/1902.08226v2 |
| https://arxiv.org/pdf/1902.08226v2.pdf | |
| PWC | https://paperswithcode.com/paper/graph-adversarial-training-dynamically |
| Repo | https://github.com/fulifeng/GraphAT |
| Framework | tf |
Region Mutual Information Loss for Semantic Segmentation
| Title | Region Mutual Information Loss for Semantic Segmentation |
| Authors | Shuai Zhao, Yang Wang, Zheng Yang, Deng Cai |
| Abstract | Semantic segmentation is a fundamental problem in computer vision. It is considered as a pixel-wise classification problem in practice, and most segmentation models use a pixel-wise loss as their optimization riterion. However, the pixel-wise loss ignores the dependencies between pixels in an image. Several ways to exploit the relationship between pixels have been investigated, \eg, conditional random fields (CRF) and pixel affinity based methods. Nevertheless, these methods usually require additional model branches, large extra memories, or more inference time. In this paper, we develop a region mutual information (RMI) loss to model the dependencies among pixels more simply and efficiently. In contrast to the pixel-wise loss which treats the pixels as independent samples, RMI uses one pixel and its neighbour pixels to represent this pixel. Then for each pixel in an image, we get a multi-dimensional point that encodes the relationship between pixels, and the image is cast into a multi-dimensional distribution of these high-dimensional points. The prediction and ground truth thus can achieve high order consistency through maximizing the mutual information (MI) between their multi-dimensional distributions. Moreover, as the actual value of the MI is hard to calculate, we derive a lower bound of the MI and maximize the lower bound to maximize the real value of the MI. RMI only requires a few extra computational resources in the training stage, and there is no overhead during testing. Experimental results demonstrate that RMI can achieve substantial and consistent improvements in performance on PASCAL VOC 2012 and CamVid datasets. The code is available at https://github.com/ZJULearning/RMI. |
| Tasks | Semantic Segmentation |
| Published | 2019-10-26 |
| URL | https://arxiv.org/abs/1910.12037v1 |
| https://arxiv.org/pdf/1910.12037v1.pdf | |
| PWC | https://paperswithcode.com/paper/region-mutual-information-loss-for-semantic |
| Repo | https://github.com/ZJULearning/RMI |
| Framework | pytorch |
GazeCorrection:Self-Guided Eye Manipulation in the wild using Self-Supervised Generative Adversarial Networks
| Title | GazeCorrection:Self-Guided Eye Manipulation in the wild using Self-Supervised Generative Adversarial Networks |
| Authors | Jichao Zhang, Meng Sun, Jingjing Chen, Hao Tang, Yan Yan, Xueying Qin, Nicu Sebe |
| Abstract | Gaze correction aims to redirect the person’s gaze into the camera by manipulating the eye region, and it can be considered as a specific image resynthesis problem. Gaze correction has a wide range of applications in real life, such as taking a picture with staring at the camera. In this paper, we propose a novel method that is based on the inpainting model to learn from the face image to fill in the missing eye regions with new contents representing corrected eye gaze. Moreover, our model does not require the training dataset labeled with the specific head pose and eye angle information, thus, the training data is easy to collect. To retain the identity information of the eye region in the original input, we propose a self-guided pretrained model to learn the angle-invariance feature. Experiments show our model achieves very compelling gaze-corrected results in the wild dataset which is collected from the website and will be introduced in details. Code is available at https://github.com/zhangqianhui/GazeCorrection. |
| Tasks | |
| Published | 2019-06-03 |
| URL | https://arxiv.org/abs/1906.00805v1 |
| https://arxiv.org/pdf/1906.00805v1.pdf | |
| PWC | https://paperswithcode.com/paper/190600805 |
| Repo | https://github.com/zhangqianhui/GazeCorrection |
| Framework | tf |
Probing Neural Network Comprehension of Natural Language Arguments
| Title | Probing Neural Network Comprehension of Natural Language Arguments |
| Authors | Timothy Niven, Hung-Yu Kao |
| Abstract | We are surprised to find that BERT’s peak performance of 77% on the Argument Reasoning Comprehension Task reaches just three points below the average untrained human baseline. However, we show that this result is entirely accounted for by exploitation of spurious statistical cues in the dataset. We analyze the nature of these cues and demonstrate that a range of models all exploit them. This analysis informs the construction of an adversarial dataset on which all models achieve random accuracy. Our adversarial dataset provides a more robust assessment of argument comprehension and should be adopted as the standard in future work. |
| Tasks | |
| Published | 2019-07-17 |
| URL | https://arxiv.org/abs/1907.07355v2 |
| https://arxiv.org/pdf/1907.07355v2.pdf | |
| PWC | https://paperswithcode.com/paper/probing-neural-network-comprehension-of |
| Repo | https://github.com/IKMLab/arct2 |
| Framework | pytorch |
Reparameterizable Subset Sampling via Continuous Relaxations
| Title | Reparameterizable Subset Sampling via Continuous Relaxations |
| Authors | Sang Michael Xie, Stefano Ermon |
| Abstract | Many machine learning tasks require sampling a subset of items from a collection based on a parameterized distribution. The Gumbel-softmax trick can be used to sample a single item, and allows for low-variance reparameterized gradients with respect to the parameters of the underlying distribution. However, stochastic optimization involving subset sampling is typically not reparameterizable. To overcome this limitation, we define a continuous relaxation of subset sampling that provides reparameterization gradients by generalizing the Gumbel-max trick. We use this approach to sample subsets of features in an instance-wise feature selection task for model interpretability, subsets of neighbors to implement a deep stochastic k-nearest neighbors model, and sub-sequences of neighbors to implement parametric t-SNE by directly comparing the identities of local neighbors. We improve performance in all these tasks by incorporating subset sampling in end-to-end training. |
| Tasks | Feature Selection, Stochastic Optimization |
| Published | 2019-01-29 |
| URL | https://arxiv.org/abs/1901.10517v4 |
| https://arxiv.org/pdf/1901.10517v4.pdf | |
| PWC | https://paperswithcode.com/paper/reparameterizable-subset-sampling-via |
| Repo | https://github.com/ermongroup/subsets |
| Framework | pytorch |
Secure multi-party linear regression at plaintext speed
| Title | Secure multi-party linear regression at plaintext speed |
| Authors | Jonathan M. Bloom |
| Abstract | We detail distributed algorithms for scalable, secure multiparty linear regression and feature selection at essentially the same speed as plaintext regression. While the core geometric ideas are simple, the recognition of their broad utility when combined is novel. Our scheme opens the door to efficient and secure genome-wide association studies across multiple biobanks. |
| Tasks | Feature Selection |
| Published | 2019-01-28 |
| URL | https://arxiv.org/abs/1901.09531v2 |
| https://arxiv.org/pdf/1901.09531v2.pdf | |
| PWC | https://paperswithcode.com/paper/secure-multi-party-linear-regression-at |
| Repo | https://github.com/jbloom22/DASH |
| Framework | none |
Adversarial Dropout for Recurrent Neural Networks
| Title | Adversarial Dropout for Recurrent Neural Networks |
| Authors | Sungrae Park, Kyungwoo Song, Mingi Ji, Wonsung Lee, Il-Chul Moon |
| Abstract | Successful application processing sequential data, such as text and speech, requires an improved generalization performance of recurrent neural networks (RNNs). Dropout techniques for RNNs were introduced to respond to these demands, but we conjecture that the dropout on RNNs could have been improved by adopting the adversarial concept. This paper investigates ways to improve the dropout for RNNs by utilizing intentionally generated dropout masks. Specifically, the guided dropout used in this research is called as adversarial dropout, which adversarially disconnects neurons that are dominantly used to predict correct targets over time. Our analysis showed that our regularizer, which consists of a gap between the original and the reconfigured RNNs, was the upper bound of the gap between the training and the inference phases of the random dropout. We demonstrated that minimizing our regularizer improved the effectiveness of the dropout for RNNs on sequential MNIST tasks, semi-supervised text classification tasks, and language modeling tasks. |
| Tasks | Language Modelling, Text Classification |
| Published | 2019-04-22 |
| URL | http://arxiv.org/abs/1904.09816v1 |
| http://arxiv.org/pdf/1904.09816v1.pdf | |
| PWC | https://paperswithcode.com/paper/adversarial-dropout-for-recurrent-neural |
| Repo | https://github.com/sungraepark/adversarial_dropout_text_classification |
| Framework | tf |
Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands
| Title | Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands |
| Authors | Hans Pinckaers, Geert Litjens |
| Abstract | Automated medical image segmentation plays a key role in quantitative research and diagnostics. Convolutional neural networks based on the U-Net architecture are the state-of-the-art. A key disadvantage is the hard-coding of the receptive field size, which requires architecture optimization for each segmentation task. Furthermore, increasing the receptive field results in an increasing number of weights. Recently, Neural Ordinary Differential Equations (NODE) have been proposed, a new type of continuous depth deep neural network. This framework allows for a dynamic receptive field at a fixed memory cost and a smaller amount of parameters. We show on a colon gland segmentation dataset (GlaS) that these NODEs can be used within the U-Net framework to improve segmentation results while reducing memory load and parameter counts. |
| Tasks | Medical Image Segmentation, Semantic Segmentation |
| Published | 2019-10-23 |
| URL | https://arxiv.org/abs/1910.10470v1 |
| https://arxiv.org/pdf/1910.10470v1.pdf | |
| PWC | https://paperswithcode.com/paper/neural-ordinary-differential-equations-for |
| Repo | https://github.com/DIAGNijmegen/neural-odes-segmentation |
| Framework | pytorch |
Concrete Autoencoders for Differentiable Feature Selection and Reconstruction
| Title | Concrete Autoencoders for Differentiable Feature Selection and Reconstruction |
| Authors | Abubakar Abid, Muhammad Fatih Balin, James Zou |
| Abstract | We introduce the concrete autoencoder, an end-to-end differentiable method for global feature selection, which efficiently identifies a subset of the most informative features and simultaneously learns a neural network to reconstruct the input data from the selected features. Our method is unsupervised, and is based on using a concrete selector layer as the encoder and using a standard neural network as the decoder. During the training phase, the temperature of the concrete selector layer is gradually decreased, which encourages a user-specified number of discrete features to be learned. During test time, the selected features can be used with the decoder network to reconstruct the remaining input features. We evaluate concrete autoencoders on a variety of datasets, where they significantly outperform state-of-the-art methods for feature selection and data reconstruction. In particular, on a large-scale gene expression dataset, the concrete autoencoder selects a small subset of genes whose expression levels can be use to impute the expression levels of the remaining genes. In doing so, it improves on the current widely-used expert-curated L1000 landmark genes, potentially reducing measurement costs by 20%. The concrete autoencoder can be implemented by adding just a few lines of code to a standard autoencoder. |
| Tasks | Feature Selection |
| Published | 2019-01-27 |
| URL | http://arxiv.org/abs/1901.09346v2 |
| http://arxiv.org/pdf/1901.09346v2.pdf | |
| PWC | https://paperswithcode.com/paper/concrete-autoencoders-for-differentiable |
| Repo | https://github.com/mfbalin/Concrete-Autoencoders |
| Framework | none |
PIPPS: Flexible Model-Based Policy Search Robust to the Curse of Chaos
| Title | PIPPS: Flexible Model-Based Policy Search Robust to the Curse of Chaos |
| Authors | Paavo Parmas, Carl Edward Rasmussen, Jan Peters, Kenji Doya |
| Abstract | Previously, the exploding gradient problem has been explained to be central in deep learning and model-based reinforcement learning, because it causes numerical issues and instability in optimization. Our experiments in model-based reinforcement learning imply that the problem is not just a numerical issue, but it may be caused by a fundamental chaos-like nature of long chains of nonlinear computations. Not only do the magnitudes of the gradients become large, the direction of the gradients becomes essentially random. We show that reparameterization gradients suffer from the problem, while likelihood ratio gradients are robust. Using our insights, we develop a model-based policy search framework, Probabilistic Inference for Particle-Based Policy Search (PIPPS), which is easily extensible, and allows for almost arbitrary models and policies, while simultaneously matching the performance of previous data-efficient learning algorithms. Finally, we invent the total propagation algorithm, which efficiently computes a union over all pathwise derivative depths during a single backwards pass, automatically giving greater weight to estimators with lower variance, sometimes improving over reparameterization gradients by $10^6$ times. |
| Tasks | |
| Published | 2019-02-04 |
| URL | http://arxiv.org/abs/1902.01240v1 |
| http://arxiv.org/pdf/1902.01240v1.pdf | |
| PWC | https://paperswithcode.com/paper/pipps-flexible-model-based-policy-search |
| Repo | https://github.com/natolambert/dynamics-learn |
| Framework | pytorch |
Sparse evolutionary Deep Learning with over one million artificial neurons on commodity hardware
| Title | Sparse evolutionary Deep Learning with over one million artificial neurons on commodity hardware |
| Authors | Shiwei Liu, Decebal Constantin Mocanu, Amarsagar Reddy Ramapuram Matavalam, Yulong Pei, Mykola Pechenizkiy |
| Abstract | Microarray gene expression has widely attracted the eyes of the public as an efficient tool for cancer diagnosis and classification. However, the very-high dimensionality and the small number of samples make it difficult for traditional machine learning algorithms to address this problem due to the high amount of computations required and overfitting. So far, the existing approaches of processing microarray datasets are still far from satisfactory and they employ two phases, feature selection (or extraction) followed by a machine learning algorithm. In this paper, we show that MultiLayer Perceptrons (MLPs) with adaptive sparse connectivity can directly handle this problem without features selection. Tested on four datasets, our novel results demonstrate that deep learning methods can be applied directly also to high dimensional non-grid like data, while learning from a small amount of labeled examples with imbalanced classes and achieving better accuracy than the traditional two phases approach. Moreover, we have been able to create sparse MLP models with over one million neurons and to train them on a typical laptop without GPU. This is with two orders of magnitude more than the largest MLPs which can run currently on commodity hardware. |
| Tasks | Feature Selection |
| Published | 2019-01-26 |
| URL | http://arxiv.org/abs/1901.09181v1 |
| http://arxiv.org/pdf/1901.09181v1.pdf | |
| PWC | https://paperswithcode.com/paper/sparse-evolutionary-deep-learning-with-over |
| Repo | https://github.com/dcmocanu/sparse-evolutionary-artificial-neural-networks |
| Framework | tf |
Improving Coordination in Small-Scale Multi-Agent Deep Reinforcement Learning through Memory-driven Communication
| Title | Improving Coordination in Small-Scale Multi-Agent Deep Reinforcement Learning through Memory-driven Communication |
| Authors | Emanuele Pesce, Giovanni Montana |
| Abstract | Deep reinforcement learning algorithms have recently been used to train multiple interacting agents in a centralised manner whilst keeping their execution decentralised. When the agents can only acquire partial observations and are faced with tasks requiring coordination and synchronisation skills, inter-agent communication plays an essential role. In this work, we propose a framework for multi-agent training using deep deterministic policy gradients that enables concurrent, end-to-end learning of an explicit communication protocol through a memory device. During training, the agents learn to perform read and write operations enabling them to infer a shared representation of the world. We empirically demonstrate that concurrent learning of the communication device and individual policies can improve inter-agent coordination and performance in small-scale systems. Our experimental results show that the proposed method achieves superior performance in scenarios with up to six agents. We illustrate how different communication patterns can emerge on six different tasks of increasing complexity. Furthermore, we study the effects of corrupting the communication channel, provide a visualisation of the time-varying memory content as the underlying task is being solved and validate the building blocks of the proposed memory device through ablation studies. |
| Tasks | |
| Published | 2019-01-12 |
| URL | https://arxiv.org/abs/1901.03887v3 |
| https://arxiv.org/pdf/1901.03887v3.pdf | |
| PWC | https://paperswithcode.com/paper/improving-coordination-in-multi-agent-deep |
| Repo | https://github.com/murtazarang/MD-MADDPG |
| Framework | tf |
Learning robust control for LQR systems with multiplicative noise via policy gradient
| Title | Learning robust control for LQR systems with multiplicative noise via policy gradient |
| Authors | Benjamin Gravell, Peyman Mohajerin Esfahani, Tyler Summers |
| Abstract | The linear quadratic regulator (LQR) problem has reemerged as an important theoretical benchmark for reinforcement learning-based control of complex dynamical systems with continuous state and action spaces. In contrast with nearly all recent work in this area, we consider multiplicative noise models, which are increasingly relevant because they explicitly incorporate inherent uncertainty and variation in the system dynamics and thereby improve robustness properties of the controller. Robustness is a critical and poorly understood issue in reinforcement learning; existing methods which do not account for uncertainty can converge to fragile policies or fail to converge at all. Additionally, intentional injection of multiplicative noise into learning algorithms can enhance robustness of policies, as observed in ad hoc work on domain randomization. Although policy gradient algorithms require optimization of a non-convex cost function, we show that the multiplicative noise LQR cost has a special property called gradient domination, which is exploited to prove global convergence of policy gradient algorithms to the globally optimum control policy with polynomial dependence on problem parameters. Results are provided both in the model-known and model-unknown settings where samples of system trajectories are used to estimate policy gradients. |
| Tasks | |
| Published | 2019-05-28 |
| URL | https://arxiv.org/abs/1905.13547v2 |
| https://arxiv.org/pdf/1905.13547v2.pdf | |
| PWC | https://paperswithcode.com/paper/190513547 |
| Repo | https://github.com/TSummersLab/polgrad-multinoise |
| Framework | none |
Exploration via Hindsight Goal Generation
| Title | Exploration via Hindsight Goal Generation |
| Authors | Zhizhou Ren, Kefan Dong, Yuan Zhou, Qiang Liu, Jian Peng |
| Abstract | Goal-oriented reinforcement learning has recently been a practical framework for robotic manipulation tasks, in which an agent is required to reach a certain goal defined by a function on the state space. However, the sparsity of such reward definition makes traditional reinforcement learning algorithms very inefficient. Hindsight Experience Replay (HER), a recent advance, has greatly improved sample efficiency and practical applicability for such problems. It exploits previous replays by constructing imaginary goals in a simple heuristic way, acting like an implicit curriculum to alleviate the challenge of sparse reward signal. In this paper, we introduce Hindsight Goal Generation (HGG), a novel algorithmic framework that generates valuable hindsight goals which are easy for an agent to achieve in the short term and are also potential for guiding the agent to reach the actual goal in the long term. We have extensively evaluated our goal generation algorithm on a number of robotic manipulation tasks and demonstrated substantially improvement over the original HER in terms of sample efficiency. |
| Tasks | |
| Published | 2019-06-10 |
| URL | https://arxiv.org/abs/1906.04279v3 |
| https://arxiv.org/pdf/1906.04279v3.pdf | |
| PWC | https://paperswithcode.com/paper/exploration-via-hindsight-goal-generation |
| Repo | https://github.com/Stilwell-Git/Hindsight-Goal-Generation |
| Framework | tf |
Every child should have parents: a taxonomy refinement algorithm based on hyperbolic term embeddings
| Title | Every child should have parents: a taxonomy refinement algorithm based on hyperbolic term embeddings |
| Authors | Rami Aly, Shantanu Acharya, Alexander Ossa, Arne Köhn, Chris Biemann, Alexander Panchenko |
| Abstract | We introduce the use of Poincar'e embeddings to improve existing state-of-the-art approaches to domain-specific taxonomy induction from text as a signal for both relocating wrong hyponym terms within a (pre-induced) taxonomy as well as for attaching disconnected terms in a taxonomy. This method substantially improves previous state-of-the-art results on the SemEval-2016 Task 13 on taxonomy extraction. We demonstrate the superiority of Poincar'e embeddings over distributional semantic representations, supporting the hypothesis that they can better capture hierarchical lexical-semantic relationships than embeddings in the Euclidean space. |
| Tasks | |
| Published | 2019-06-05 |
| URL | https://arxiv.org/abs/1906.02002v1 |
| https://arxiv.org/pdf/1906.02002v1.pdf | |
| PWC | https://paperswithcode.com/paper/every-child-should-have-parents-a-taxonomy |
| Repo | https://github.com/uhh-lt/Taxonomy_Refinement_Embeddings |
| Framework | none |
