Paper Group NANR 72
Efficacy of Pixel-Level OOD Detection for Semantic Segmentation. Customizing Sequence Generation with Multi-Task Dynamical Systems. Learning Representations in Reinforcement Learning: an Information Bottleneck Approach. RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers. LEX-GAN: Layered Explainable Rumor Detector Based on …
Efficacy of Pixel-Level OOD Detection for Semantic Segmentation
| Title | Efficacy of Pixel-Level OOD Detection for Semantic Segmentation |
| Authors | Anonymous |
| Abstract | The detection of out of distribution samples for image classification has been widely researched. Safety critical applications, such as autonomous driving, would benefit from the ability to localise the unusual objects causing the image to be out of distribution. This paper adapts state-of-the-art methods for detecting out of distribution images for image classification to the new task of detecting out of distribution pixels, which can localise the unusual objects. It further experimentally compares the adapted methods on two new datasets derived from existing semantic segmentation datasets using PSPNet and DeeplabV3+ architectures, as well as proposing a new metric for the task. The evaluation shows that the performance ranking of the compared methods does not transfer to the new task and every method performs significantly worse than their image-level counterparts. |
| Tasks | Autonomous Driving, Image Classification, Semantic Segmentation |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=SJlxglSFPB |
| https://openreview.net/pdf?id=SJlxglSFPB | |
| PWC | https://paperswithcode.com/paper/efficacy-of-pixel-level-ood-detection-for |
| Repo | |
| Framework | |
Customizing Sequence Generation with Multi-Task Dynamical Systems
| Title | Customizing Sequence Generation with Multi-Task Dynamical Systems |
| Authors | Anonymous |
| Abstract | Dynamical system models (including RNNs) often lack the ability to adapt the sequence generation or prediction to a given context, limiting their real-world application. In this paper we show that hierarchical multi-task dynamical systems (MTDSs) provide direct user control over sequence generation, via use of a latent code z that specifies the customization to the individual data sequence. This enables style transfer, interpolation and morphing within generated sequences. We show the MTDS can improve predictions via latent code interpolation, and avoid the long-term performance degradation of standard RNN approaches. |
| Tasks | Style Transfer |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=Bkln2a4tPB |
| https://openreview.net/pdf?id=Bkln2a4tPB | |
| PWC | https://paperswithcode.com/paper/customizing-sequence-generation-with-multi-1 |
| Repo | |
| Framework | |
Learning Representations in Reinforcement Learning: an Information Bottleneck Approach
| Title | Learning Representations in Reinforcement Learning: an Information Bottleneck Approach |
| Authors | Yingjun Pei, Xinwen Hou |
| Abstract | The information bottleneck principle is an elegant and useful approach to representation learning. In this paper, we investigate the problem of representation learning in the context of reinforcement learning using the information bottleneck framework, aiming at improving the sample efficiency of the learning algorithms.We analytically derive the optimal conditional distribution of the representation, and provide a variational lower bound. Then, we maximize this lower bound with the Stein variational (SV) gradient method. We incorporate this framework in the advantageous actor critic algorithm (A2C) and the proximal policy optimization algorithm (PPO). Our experimental results show that our framework can improve the sample efficiency of vanilla A2C and PPO significantly. Finally, we study the information-bottleneck (IB) perspective in deep RL with the algorithm called mutual information neural estimation(MINE). We experimentally verify that the information extraction-compression process also exists in deep RL and our framework is capable of accelerating this process. We also analyze the relationship between MINE and our method, through this relationship, we theoretically derive an algorithm to optimize our IB framework without constructing the lower bound. |
| Tasks | Representation Learning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=Syl-xpNtwS |
| https://openreview.net/pdf?id=Syl-xpNtwS | |
| PWC | https://paperswithcode.com/paper/learning-representations-in-reinforcement-1 |
| Repo | |
| Framework | |
RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers
| Title | RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers |
| Authors | Anonymous |
| Abstract | When translating natural language questions into SQL queries to answer questions from a database, contemporary semantic parsing models struggle to generalize to unseen database schemas. The generalization challenge lies in (a) encoding the database relations in an accessible way for the semantic parser, and (b) modeling alignment between database columns and their mentions in a given query. We present a unified framework, based on the relation-aware self-attention mechanism,to address schema encoding, schema linking, and feature representation within a text-to-SQL encoder. On the challenging Spider dataset this framework boosts the exact match accuracy to 53.7%, compared to 47.4% for the previous state-of-the-art model unaugmented with BERT embeddings. In addition, we observe qualitative improvements in the model’s understanding of schema linking and alignment. |
| Tasks | Semantic Parsing, Text-To-Sql |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=H1egcgHtvB |
| https://openreview.net/pdf?id=H1egcgHtvB | |
| PWC | https://paperswithcode.com/paper/rat-sql-relation-aware-schema-encoding-and |
| Repo | |
| Framework | |
LEX-GAN: Layered Explainable Rumor Detector Based on Generative Adversarial Networks
| Title | LEX-GAN: Layered Explainable Rumor Detector Based on Generative Adversarial Networks |
| Authors | Anonymous |
| Abstract | Social media have emerged to be increasingly popular and have been used as tools for gathering and propagating information. However, the vigorous growth of social media contributes to the fast-spreading and far-reaching rumors. Rumor detection has become a necessary defense. Traditional rumor detection methods based on hand-crafted feature selection are replaced by automatic approaches that are based on Artificial Intelligence (AI). AI decision making systems need to have the necessary means, such as explainability to assure users their trustworthiness. Inspired by the thriving development of Generative Adversarial Networks (GANs) on text applications, we propose LEX-GAN, a GAN-based layered explainable rumor detector to improve the detection quality and provide explainability. Unlike fake news detection that needs a previously collected verified news database, LEX-GAN realizes explainable rumor detection based on only tweet-level text. LEX-GAN is trained with generated non-rumor-looking rumors. The generators produce rumors by intelligently inserting controversial information in non-rumors, and force the discriminators to detect detailed glitches and deduce exactly which parts in the sentence are problematic. The layered structures in both generative and discriminative model contributes to the high performance. We show LEX-GAN’s mutation detection ability in textural sequences by performing a gene classification and mutation detection task. |
| Tasks | Decision Making, Fake News Detection, Feature Selection |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=S1lukyrKPr |
| https://openreview.net/pdf?id=S1lukyrKPr | |
| PWC | https://paperswithcode.com/paper/lex-gan-layered-explainable-rumor-detector |
| Repo | |
| Framework | |
The Usual Suspects? Reassessing Blame for VAE Posterior Collapse
| Title | The Usual Suspects? Reassessing Blame for VAE Posterior Collapse |
| Authors | Anonymous |
| Abstract | In narrow asymptotic settings Gaussian VAE models of continuous data have been shown to possess global optima aligned with ground-truth distributions. Even so, it is well known that poor solutions whereby the latent posterior collapses to an uninformative prior are sometimes obtained in practice. However, contrary to conventional wisdom that largely assigns blame for this phenomena on the undue influence of KL-divergence regularization, we will argue that posterior collapse is, at least in part, a direct consequence of bad local minima inherent to the loss surface of deep autoencoder networks. In particular, we prove that even small nonlinear perturbations of affine VAE decoder models can produce such minima, and in deeper models, analogous minima can force the VAE to behave like an aggressive truncation operator, provably discarding information along all latent dimensions in certain circumstances. Regardless, the underlying message here is not meant to undercut valuable existing explanations of posterior collapse, but rather, to refine the discussion and elucidate alternative risk factors that may have been previously underappreciated. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=r1lIKlSYvH |
| https://openreview.net/pdf?id=r1lIKlSYvH | |
| PWC | https://paperswithcode.com/paper/the-usual-suspects-reassessing-blame-for-vae |
| Repo | |
| Framework | |
Generative Restricted Kernel Machines
| Title | Generative Restricted Kernel Machines |
| Authors | Anonymous |
| Abstract | We introduce a novel framework for generative models based on Restricted Kernel Machines (RKMs) with multi-view generation and uncorrelated feature learning capabilities, called Gen-RKM. To incorporate multi-view generation, this mechanism uses a shared representation of data from various views. The mechanism is flexible to incorporate both kernel-based, (deep) neural network and convolutional based models within the same setting. To update the parameters of the network, we propose a novel training procedure which jointly learns the features and shared representation. Experiments demonstrate the potential of the framework through qualitative evaluation of generated samples. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=ryghPCVYvH |
| https://openreview.net/pdf?id=ryghPCVYvH | |
| PWC | https://paperswithcode.com/paper/generative-restricted-kernel-machines-1 |
| Repo | |
| Framework | |
Which Tasks Should Be Learned Together in Multi-task Learning?
| Title | Which Tasks Should Be Learned Together in Multi-task Learning? |
| Authors | Anonymous |
| Abstract | Many computer vision applications require solving multiple tasks in real-time. A neural network can be trained to solve multiple tasks simultaneously using ‘multi-task learning’. This saves computation at inference time as only a single network needs to be evaluated. Unfortunately, this often leads to inferior overall performance as task objectives compete, which consequently poses the question: which tasks should and should not be learned together in one network when employing multi-task learning? We systematically study task cooperation and competition and propose a framework for assigning tasks to a few neural networks such that cooperating tasks are computed by the same neural network, while competing tasks are computed by different networks. Our framework offers a time-accuracy trade-off and can produce better accuracy using less inference time than not only a single large multi-task neural network but also many single-task networks. |
| Tasks | Multi-Task Learning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=HJlTpCEKvS |
| https://openreview.net/pdf?id=HJlTpCEKvS | |
| PWC | https://paperswithcode.com/paper/which-tasks-should-be-learned-together-in-1 |
| Repo | |
| Framework | |
Infinite-horizon Off-Policy Policy Evaluation with Multiple Behavior Policies
| Title | Infinite-horizon Off-Policy Policy Evaluation with Multiple Behavior Policies |
| Authors | Anonymous |
| Abstract | We consider off-policy policy evaluation when the trajectory data are generated by multiple behavior policies. Recent work has shown the key role played by the state or state-action stationary distribution corrections in the infinite horizon context for off-policy policy evaluation. We propose estimated mixture policy (EMP), a novel class of partially policy-agnostic methods to accurately estimate those quantities. With careful analysis, we show that EMP gives rise to estimates with reduced variance for estimating the state stationary distribution correction while it also offers a useful induction bias for estimating the state-action stationary distribution correction. In extensive experiments with both continuous and discrete environments, we demonstrate that our algorithm offers significantly improved accuracy compared to the state-of-the-art methods. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rkgU1gHtvr |
| https://openreview.net/pdf?id=rkgU1gHtvr | |
| PWC | https://paperswithcode.com/paper/infinite-horizon-off-policy-policy-evaluation |
| Repo | |
| Framework | |
Continuous Meta-Learning without Tasks
| Title | Continuous Meta-Learning without Tasks |
| Authors | Anonymous |
| Abstract | Meta-learning is a promising strategy for learning to efficiently learn within new tasks, using data gathered from a distribution of tasks. However, the meta-learning literature thus far has focused on the task segmented setting, where at train-time, offline data is assumed to be split according to the underlying task, and at test-time, the algorithms are optimized to learn in a single task. In this work, we enable the application of generic meta-learning algorithms to settings where this task segmentation is unavailable, such as continual online learning with a time-varying task. We present meta-learning via online changepoint analysis (MOCA), an approach which augments a meta-learning algorithm with a differentiable Bayesian changepoint detection scheme. The framework allows both training and testing directly on time series data without segmenting it into discrete tasks. We demonstrate the utility of this approach on a nonlinear meta-regression benchmark as well as two meta-image-classification benchmarks. |
| Tasks | Image Classification, Meta-Learning, Time Series |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=r1l1myStwr |
| https://openreview.net/pdf?id=r1l1myStwr | |
| PWC | https://paperswithcode.com/paper/continuous-meta-learning-without-tasks |
| Repo | |
| Framework | |
FAST LEARNING VIA EPISODIC MEMORY: A PERSPECTIVE FROM ANIMAL DECISION-MAKING
| Title | FAST LEARNING VIA EPISODIC MEMORY: A PERSPECTIVE FROM ANIMAL DECISION-MAKING |
| Authors | Anonymous |
| Abstract | A typical experiment to study cognitive function is to train animals to perform tasks, while the researcher records the electrical activity of the animals neurons. The main obstacle faced, when using this type of electrophysiological experiment to uncover the circuit mechanisms underlying complex behaviors, is our incomplete access to relevant circuits in the brain. One promising approach is to model neural circuits using an artificial neural network (ANN), which can provide complete access to the “neural circuits” responsible for a behavior. More recently, reinforcement learning models have been adopted to understand the functions of cortico-basal ganglia circuits as reward-based learning has been found in mammalian brain. In this paper, we propose a Biologically-plausible Actor-Critic with Episodic Memory (B-ACEM) framework to model a prefrontal cortex-basal ganglia-hippocampus (PFC-BG) circuit, which is verified to capture the behavioral findings from a well-known perceptual decision-making task, i.e., random dots motion discrimination. This B-ACEM framework links neural computation to behaviors, on which we can explore how episodic memory should be considered to govern future decision. Experiments are conducted using different settings of the episodic memory and results show that all patterns of episodic memories can speed up learning. In particular, salient events are prioritized to propagate reward information and guide decisions. Our B-ACEM framework and the built-on experiments give inspirations to both designs for more standard decision-making models in biological system and a more biologically-plausible ANN. |
| Tasks | Decision Making |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=BJl4pA4Kwr |
| https://openreview.net/pdf?id=BJl4pA4Kwr | |
| PWC | https://paperswithcode.com/paper/fast-learning-via-episodic-memory-a |
| Repo | |
| Framework | |
Implicit Generative Modeling for Efficient Exploration
| Title | Implicit Generative Modeling for Efficient Exploration |
| Authors | Anonymous |
| Abstract | Efficient exploration remains a challenging problem in reinforcement learning, especially for those tasks where rewards from environments are sparse. A commonly used approach for exploring such environments is to introduce some “intrinsic” reward. In this work, we focus on model uncertainty estimation as an intrinsic reward for efficient exploration. In particular, we introduce an implicit generative modeling approach to estimate a Bayesian uncertainty of the agent’s belief of the environment dynamics. Each random draw from our generative model is a neural network that instantiates the dynamic function, hence multiple draws would approximate the posterior, and the variance in the future prediction based on this posterior is used as an intrinsic reward for exploration. We design a training algorithm for our generative model based on the amortized Stein Variational Gradient Descent. In experiments, we compare our implementation with state-of-the-art intrinsic reward-based exploration approaches, including two recent approaches based on an ensemble of dynamic models. In challenging exploration tasks, our implicit generative model consistently outperforms competing approaches regarding data efficiency in exploration. |
| Tasks | Efficient Exploration, Future prediction |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=BkgeQ1BYwS |
| https://openreview.net/pdf?id=BkgeQ1BYwS | |
| PWC | https://paperswithcode.com/paper/implicit-generative-modeling-for-efficient |
| Repo | |
| Framework | |
Selfish Emergent Communication
| Title | Selfish Emergent Communication |
| Authors | Anonymous |
| Abstract | Current literature in machine learning holds that unaligned, self-interested agents do not learn to use an emergent communication channel. We introduce a new sender-receiver game to study emergent communication for this spectrum of partially-competitive scenarios and put special care into evaluation. We find that communication can indeed emerge in partially-competitive scenarios, and we discover three things that are tied to improving it. First, that selfish communication is proportional to cooperation, and it naturally occurs for situations that are more cooperative than competitive. Second, that stability and performance are improved by using LOLA (Foerster et al, 2018), especially in more competitive scenarios. And third, that discrete protocols lend themselves better to learning cooperative communication than continuous ones. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=B1liIlBKvS |
| https://openreview.net/pdf?id=B1liIlBKvS | |
| PWC | https://paperswithcode.com/paper/selfish-emergent-communication |
| Repo | |
| Framework | |
Robust Graph Representation Learning via Neural Sparsification
| Title | Robust Graph Representation Learning via Neural Sparsification |
| Authors | Anonymous |
| Abstract | Graph representation learning serves as the core of many important prediction tasks, ranging from product recommendation in online marketing to fraud detection in financial domain. Real-life graphs are usually large with complex local neighborhood, where each node is described by a rich set of features and easily connects to dozens or even hundreds of neighbors. Most existing graph learning techniques rely on neighborhood aggregation, however, the complexity on real-life graphs is usually high, posing non-trivial overfitting risk during model training. In this paper, we present Neural Sparsification (NeuralSparse), a supervised graph sparsification technique that mitigates the overfitting risk by reducing the complexity of input graphs. Our method takes both structural and non-structural information as input, utilizes deep neural networks to parameterize the sparsification process, and optimizes the parameters by feedback signals from downstream tasks. Under the NeuralSparse framework, supervised graph sparsification could seamlessly connect with existing graph neural networks for more robust performance on testing data. Experimental results on both benchmark and private datasets show that NeuralSparse can effectively improve testing accuracy and bring up to 7.4% improvement when working with existing graph neural networks on node classification tasks. |
| Tasks | Fraud Detection, Graph Representation Learning, Node Classification, Product Recommendation, Representation Learning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=S1emOTNKvS |
| https://openreview.net/pdf?id=S1emOTNKvS | |
| PWC | https://paperswithcode.com/paper/robust-graph-representation-learning-via |
| Repo | |
| Framework | |
NEURAL EXECUTION ENGINES
| Title | NEURAL EXECUTION ENGINES |
| Authors | Yujun Yan, Kevin Swersky, Danai Koutra, Parthasarathy Ranganathan, Milad Hashemi |
| Abstract | Turing complete computation and reasoning are often regarded as necessary pre- cursors to general intelligence. There has been a significant body of work studying neural networks that mimic general computation, but these networks fail to generalize to data distributions that are outside of their training set. We study this problem through the lens of fundamental computer science problems: sorting and graph processing. We modify the masking mechanism of a transformer in order to allow them to implement rudimentary functions with strong generalization. We call this model the Neural Execution Engine, and show that it learns, through supervision, to numerically compute the basic subroutines comprising these algorithms with near perfect accuracy. Moreover, it retains this level of accuracy while generalizing to unseen data and long sequences outside of the training distribution. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rJg7BA4YDr |
| https://openreview.net/pdf?id=rJg7BA4YDr | |
| PWC | https://paperswithcode.com/paper/neural-execution-engines |
| Repo | |
| Framework | |
