February 1, 2020

3081 words 15 mins read

Paper Group AWR 180

AI Feynman: a Physics-Inspired Method for Symbolic Regression. Neural Logic Machines. Enhancing Relation Extraction Using Syntactic Indicators and Sentential Contexts. BioFLAIR: Pretrained Pooled Contextualized Embeddings for Biomedical Sequence Labeling Tasks. Baconian: A Unified Opensource Framework for Model-Based Reinforcement Learning. Revisit …

AI Feynman: a Physics-Inspired Method for Symbolic Regression


Title	AI Feynman: a Physics-Inspired Method for Symbolic Regression
Authors	Silviu-Marian Udrescu, Max Tegmark
Abstract	A core challenge for both physics and artificial intellicence (AI) is symbolic regression: finding a symbolic expression that matches data from an unknown function. Although this problem is likely to be NP-hard in principle, functions of practical interest often exhibit symmetries, separability, compositionality and other simplifying properties. In this spirit, we develop a recursive multidimensional symbolic regression algorithm that combines neural network fitting with a suite of physics-inspired techniques. We apply it to 100 equations from the Feynman Lectures on Physics, and it discovers all of them, while previous publicly available software cracks only 71; for a more difficult test set, we improve the state of the art success rate from 15% to 90%.
Tasks
Published	2019-05-27
URL	https://arxiv.org/abs/1905.11481v1
PDF	https://arxiv.org/pdf/1905.11481v1.pdf
PWC	https://paperswithcode.com/paper/ai-feynman-a-physics-inspired-method-for
Repo	https://github.com/SJ001/AI-Feynman
Framework	none

Neural Logic Machines


Title	Neural Logic Machines
Authors	Honghua Dong, Jiayuan Mao, Tian Lin, Chong Wang, Lihong Li, Denny Zhou
Abstract	We propose the Neural Logic Machine (NLM), a neural-symbolic architecture for both inductive learning and logic reasoning. NLMs exploit the power of both neural networks—as function approximators, and logic programming—as a symbolic processor for objects with properties, relations, logic connectives, and quantifiers. After being trained on small-scale tasks (such as sorting short arrays), NLMs can recover lifted rules, and generalize to large-scale tasks (such as sorting longer arrays). In our experiments, NLMs achieve perfect generalization in a number of tasks, from relational reasoning tasks on the family tree and general graphs, to decision making tasks including sorting arrays, finding shortest paths, and playing the blocks world. Most of these tasks are hard to accomplish for neural networks or inductive logic programming alone.
Tasks	Decision Making, Relational Reasoning
Published	2019-04-26
URL	http://arxiv.org/abs/1904.11694v1
PDF	http://arxiv.org/pdf/1904.11694v1.pdf
PWC	https://paperswithcode.com/paper/neural-logic-machines-1
Repo	https://github.com/google/neural-logic-machines
Framework	pytorch

Enhancing Relation Extraction Using Syntactic Indicators and Sentential Contexts


Title	Enhancing Relation Extraction Using Syntactic Indicators and Sentential Contexts
Authors	Qiongxing Tao, Xiangfeng Luo, Hao Wang
Abstract	State-of-the-art methods for relation extraction consider the sentential context by modeling the entire sentence. However, syntactic indicators, certain phrases or words like prepositions that are more informative than other words and may be beneficial for identifying semantic relations. Other approaches using fixed text triggers capture such information but ignore the lexical diversity. To leverage both syntactic indicators and sentential contexts, we propose an indicator-aware approach for relation extraction. Firstly, we extract syntactic indicators under the guidance of syntactic knowledge. Then we construct a neural network to incorporate both syntactic indicators and the entire sentences into better relation representations. By this way, the proposed model alleviates the impact of noisy information from entire sentences and breaks the limit of text triggers. Experiments on the SemEval-2010 Task 8 benchmark dataset show that our model significantly outperforms the state-of-the-art methods.
Tasks	Relation Extraction
Published	2019-12-04
URL	https://arxiv.org/abs/1912.01858v1
PDF	https://arxiv.org/pdf/1912.01858v1.pdf
PWC	https://paperswithcode.com/paper/enhancing-relation-extraction-using-syntactic
Repo	https://github.com/wang-h/bert-relation-classification
Framework	pytorch

BioFLAIR: Pretrained Pooled Contextualized Embeddings for Biomedical Sequence Labeling Tasks


Title	BioFLAIR: Pretrained Pooled Contextualized Embeddings for Biomedical Sequence Labeling Tasks
Authors	Shreyas Sharma, Ron Daniel Jr
Abstract	Biomedical Named Entity Recognition (NER) is a challenging problem in biomedical information processing due to the widespread ambiguity of out of context terms and extensive lexical variations. Performance on bioNER benchmarks continues to improve due to advances like BERT, GPT, and XLNet. FLAIR (1) is an alternative embedding model which is less computationally intensive than the others mentioned. We test FLAIR and its pretrained PubMed embeddings (which we term BioFLAIR) on a variety of bio NER tasks and compare those with results from BERT-type networks. We also investigate the effects of a small amount of additional pretraining on PubMed content, and of combining FLAIR and ELMO models. We find that with the provided embeddings, FLAIR performs on-par with the BERT networks - even establishing a new state of the art on one benchmark. Additional pretraining did not provide a clear benefit, although this might change with even more pretraining being done. Stacking the FLAIR embeddings with others typically does provide a boost in the benchmark results.
Tasks	Medical Named Entity Recognition, Named Entity Recognition
Published	2019-08-13
URL	https://arxiv.org/abs/1908.05760v1
PDF	https://arxiv.org/pdf/1908.05760v1.pdf
PWC	https://paperswithcode.com/paper/bioflair-pretrained-pooled-contextualized
Repo	https://github.com/shreyashub/BioFLAIR
Framework	none

Baconian: A Unified Opensource Framework for Model-Based Reinforcement Learning


Title	Baconian: A Unified Opensource Framework for Model-Based Reinforcement Learning
Authors	Linsen Dong, Guanyu Gao, Yuanlong Li, Yonggang Wen
Abstract	Model-Based Reinforcement Learning (MBRL) is one category of Reinforcement Learning (RL) methods which can improve sampling efficiency by modeling and approximating system dynamics. It has been widely adopted in the research of robotics, autonomous driving, etc. Despite its popularity, there still lacks some sophisticated and reusable opensource frameworks to facilitate MBRL research and experiments. To fill this gap, we develop a flexible and modularized framework, Baconian, which allows researchers to easily implement a MBRL testbed by customizing or building upon our provided modules and algorithms. Our framework can free the users from re-implementing popular MBRL algorithms from scratch thus greatly saves the users’ efforts.
Tasks	Autonomous Driving
Published	2019-04-23
URL	http://arxiv.org/abs/1904.10762v1
PDF	http://arxiv.org/pdf/1904.10762v1.pdf
PWC	https://paperswithcode.com/paper/baconian-a-unified-opensource-framework-for
Repo	https://github.com/Lukeeeeee/baconian-project
Framework	none

Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution


Title	Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution
Authors	Shany Barhom, Vered Shwartz, Alon Eirew, Michael Bugert, Nils Reimers, Ido Dagan
Abstract	Recognizing coreferring events and entities across multiple texts is crucial for many NLP applications. Despite the task’s importance, research focus was given mostly to within-document entity coreference, with rather little attention to the other variants. We propose a neural architecture for cross-document coreference resolution. Inspired by Lee et al (2012), we jointly model entity and event coreference. We represent an event (entity) mention using its lexical span, surrounding context, and relation to entity (event) mentions via predicate-arguments structures. Our model outperforms the previous state-of-the-art event coreference model on ECB+, while providing the first entity coreference results on this corpus. Our analysis confirms that all our representation elements, including the mention span itself, its context, and the relation to other mentions contribute to the model’s success.
Tasks	Coreference Resolution
Published	2019-06-04
URL	https://arxiv.org/abs/1906.01753v1
PDF	https://arxiv.org/pdf/1906.01753v1.pdf
PWC	https://paperswithcode.com/paper/revisiting-joint-modeling-of-cross-document
Repo	https://github.com/shanybar/event_entity_coref_ecb_plus
Framework	pytorch

Let’s Play Again: Variability of Deep Reinforcement Learning Agents in Atari Environments


Title	Let’s Play Again: Variability of Deep Reinforcement Learning Agents in Atari Environments
Authors	Kaleigh Clary, Emma Tosch, John Foley, David Jensen
Abstract	Reproducibility in reinforcement learning is challenging: uncontrolled stochasticity from many sources, such as the learning algorithm, the learned policy, and the environment itself have led researchers to report the performance of learned agents using aggregate metrics of performance over multiple random seeds for a single environment. Unfortunately, there are still pernicious sources of variability in reinforcement learning agents that make reporting common summary statistics an unsound metric for performance. Our experiments demonstrate the variability of common agents used in the popular OpenAI Baselines repository. We make the case for reporting post-training agent performance as a distribution, rather than a point estimate.
Tasks	Atari Games
Published	2019-04-12
URL	http://arxiv.org/abs/1904.06312v1
PDF	http://arxiv.org/pdf/1904.06312v1.pdf
PWC	https://paperswithcode.com/paper/lets-play-again-variability-of-deep
Repo	https://github.com/kclary/variability-RL
Framework	none

Deep Policies for Width-Based Planning in Pixel Domains


Title	Deep Policies for Width-Based Planning in Pixel Domains
Authors	Miquel Junyent, Anders Jonsson, Vicenç Gómez
Abstract	Width-based planning has demonstrated great success in recent years due to its ability to scale independently of the size of the state space. For example, Bandres et al. (2018) introduced a rollout version of the Iterated Width algorithm whose performance compares well with humans and learning methods in the pixel setting of the Atari games suite. In this setting, planning is done on-line using the “screen” states and selecting actions by looking ahead into the future. However, this algorithm is purely exploratory and does not leverage past reward information. Furthermore, it requires the state to be factored into features that need to be pre-defined for the particular task, e.g., the B-PROST pixel features. In this work, we extend width-based planning by incorporating an explicit policy in the action selection mechanism. Our method, called $\pi$-IW, interleaves width-based planning and policy learning using the state-actions visited by the planner. The policy estimate takes the form of a neural network and is in turn used to guide the planning step, thus reinforcing promising paths. Surprisingly, we observe that the representation learned by the neural network can be used as a feature space for the width-based planner without degrading its performance, thus removing the requirement of pre-defined features for the planner. We compare $\pi$-IW with previous width-based methods and with AlphaZero, a method that also interleaves planning and learning, in simple environments, and show that $\pi$-IW has superior performance. We also show that $\pi$-IW algorithm outperforms previous width-based methods in the pixel setting of Atari games suite.
Tasks	Atari Games
Published	2019-04-12
URL	http://arxiv.org/abs/1904.07091v1
PDF	http://arxiv.org/pdf/1904.07091v1.pdf
PWC	https://paperswithcode.com/paper/deep-policies-for-width-based-planning-in
Repo	https://github.com/aig-upf/pi-IW
Framework	tf

Obstacle Tower: A Generalization Challenge in Vision, Control, and Planning


Title	Obstacle Tower: A Generalization Challenge in Vision, Control, and Planning
Authors	Arthur Juliani, Ahmed Khalifa, Vincent-Pierre Berges, Jonathan Harper, Ervin Teng, Hunter Henry, Adam Crespi, Julian Togelius, Danny Lange
Abstract	The rapid pace of recent research in AI has been driven in part by the presence of fast and challenging simulation environments. These environments often take the form of games; with tasks ranging from simple board games, to competitive video games. We propose a new benchmark - Obstacle Tower: a high fidelity, 3D, 3rd person, procedurally generated environment. An agent playing Obstacle Tower must learn to solve both low-level control and high-level planning problems in tandem while learning from pixels and a sparse reward signal. Unlike other benchmarks such as the Arcade Learning Environment, evaluation of agent performance in Obstacle Tower is based on an agent’s ability to perform well on unseen instances of the environment. In this paper we outline the environment and provide a set of baseline results produced by current state-of-the-art Deep RL methods as well as human players. These algorithms fail to produce agents capable of performing near human level.
Tasks	Atari Games, Board Games
Published	2019-02-04
URL	https://arxiv.org/abs/1902.01378v2
PDF	https://arxiv.org/pdf/1902.01378v2.pdf
PWC	https://paperswithcode.com/paper/obstacle-tower-a-generalization-challenge-in
Repo	https://github.com/dazcona/obstacletower
Framework	none

Combinational Q-Learning for Dou Di Zhu


Title	Combinational Q-Learning for Dou Di Zhu
Authors	Yang You, Liangwei Li, Baisong Guo, Weiming Wang, Cewu Lu
Abstract	Deep reinforcement learning (DRL) has gained a lot of attention in recent years, and has been proven to be able to play Atari games and Go at or above human levels. However, those games are assumed to have a small fixed number of actions and could be trained with a simple CNN network. In this paper, we study a special class of Asian popular card games called Dou Di Zhu, in which two adversarial groups of agents must consider numerous card combinations at each time step, leading to huge number of actions. We propose a novel method to handle combinatorial actions, which we call combinational Q-learning (CQL). We employ a two-stage network to reduce action space and also leverage order-invariant max-pooling operations to extract relationships between primitive actions. Results show that our method prevails over state-of-the art methods like naive Q-learning and A3C. We develop an easy-to-use card game environments and train all agents adversarially from sractch, with only knowledge of game rules and verify that our agents are comparative to humans. Our code to reproduce all reported results will be available online.
Tasks	Atari Games, Card Games, Q-Learning
Published	2019-01-24
URL	http://arxiv.org/abs/1901.08925v2
PDF	http://arxiv.org/pdf/1901.08925v2.pdf
PWC	https://paperswithcode.com/paper/combinational-q-learning-for-dou-di-zhu
Repo	https://github.com/qq456cvb/doudizhu-C
Framework	tf

Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEs


Title	Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEs
Authors	Nicholas Watters, Loic Matthey, Christopher P. Burgess, Alexander Lerchner
Abstract	We present a simple neural rendering architecture that helps variational autoencoders (VAEs) learn disentangled representations. Instead of the deconvolutional network typically used in the decoder of VAEs, we tile (broadcast) the latent vector across space, concatenate fixed X- and Y-“coordinate” channels, and apply a fully convolutional network with 1x1 stride. This provides an architectural prior for dissociating positional from non-positional features in the latent distribution of VAEs, yet without providing any explicit supervision to this effect. We show that this architecture, which we term the Spatial Broadcast decoder, improves disentangling, reconstruction accuracy, and generalization to held-out regions in data space. It provides a particularly dramatic benefit when applied to datasets with small objects. We also emphasize a method for visualizing learned latent spaces that helped us diagnose our models and may prove useful for others aiming to assess data representations. Finally, we show the Spatial Broadcast Decoder is complementary to state-of-the-art (SOTA) disentangling techniques and when incorporated improves their performance.
Tasks
Published	2019-01-21
URL	https://arxiv.org/abs/1901.07017v2
PDF	https://arxiv.org/pdf/1901.07017v2.pdf
PWC	https://paperswithcode.com/paper/spatial-broadcast-decoder-a-simple
Repo	https://github.com/deepmind/spriteworld
Framework	none

Zero-shot task adaptation by homoiconic meta-mapping


Title	Zero-shot task adaptation by homoiconic meta-mapping
Authors	Andrew K. Lampinen, James L. McClelland
Abstract	How can deep learning systems flexibly reuse their knowledge? Toward this goal, we propose a new class of challenges, and a class of architectures that can solve them. The challenges are meta-mappings, which involve systematically transforming task behaviors to adapt to new tasks zero-shot. The key to achieving these challenges is representing the task being performed in such a way that this task representation is itself transformable. We therefore draw inspiration from functional programming and recent work in meta-learning to propose a class of Homoiconic Meta-Mapping (HoMM) approaches that represent data points and tasks in a shared latent space, and learn to infer transformations of that space. HoMM approaches can be applied to any type of machine learning task. We demonstrate the utility of this perspective by exhibiting zero-shot remapping of behavior to adapt to new tasks.
Tasks	Meta-Learning
Published	2019-05-23
URL	https://arxiv.org/abs/1905.09950v4
PDF	https://arxiv.org/pdf/1905.09950v4.pdf
PWC	https://paperswithcode.com/paper/embedded-meta-learning-toward-more-flexible
Repo	https://github.com/lampinen/HoMM
Framework	tf

Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs using Deep Convolutional Neural Networks


Title	Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs using Deep Convolutional Neural Networks
Authors	Aleksei Tiulpin, Simo Saarakkala
Abstract	Knee osteoarthritis (OA) is the most common musculoskeletal disease in the world. In primary healthcare, knee OA is diagnosed using clinical examination and radiographic assessment. Osteoarthritis Research Society International (OARSI) atlas of OA radiographic features allows to perform independent assessment of knee osteophytes, joint space narrowing and other knee features. This provides a fine-grained OA severity assessment of the knee, compared to the gold standard and most commonly used Kellgren-Lawrence (KL) composite score. However, both OARSI and KL grading systems suffer from moderate inter-rater agreement, and therefore, the use of computer-aided methods could help to improve the reliability of the process. In this study, we developed a robust, automatic method to simultaneously predict KL and OARSI grades in knee radiographs. Our method is based on Deep Learning and leverages an ensemble of deep residual networks with 50 layers, squeeze-excitation and ResNeXt blocks. Here, we used transfer learning from ImageNet with a fine-tuning on the whole Osteoarthritis Initiative (OAI) dataset. An independent testing of our model was performed on the whole Multicenter Osteoarthritis Study (MOST) dataset. Our multi-task method yielded Cohen’s kappa coefficients of 0.82 for KL-grade and 0.79, 0.84, 0.94, 0.83, 0.84, 0.90 for femoral osteophytes, tibial osteophytes and joint space narrowing for lateral and medial compartments respectively. Furthermore, our method yielded area under the ROC curve of 0.98 and average precision of 0.98 for detecting the presence of radiographic OA (KL $\geq 2$), which is better than the current state-of-the-art.
Tasks	Transfer Learning
Published	2019-07-18
URL	https://arxiv.org/abs/1907.08020v1
PDF	https://arxiv.org/pdf/1907.08020v1.pdf
PWC	https://paperswithcode.com/paper/automatic-grading-of-individual-knee
Repo	https://github.com/MIPT-Oulu/KneeOARSIGrading
Framework	pytorch

Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset


Title	Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset
Authors	Tianyu Wang, Xin Yang, Ke Xu, Shaozhe Chen, Qiang Zhang, Rynson Lau
Abstract	Removing rain streaks from a single image has been drawing considerable attention as rain streaks can severely degrade the image quality and affect the performance of existing outdoor vision tasks. While recent CNN-based derainers have reported promising performances, deraining remains an open problem for two reasons. First, existing synthesized rain datasets have only limited realism, in terms of modeling real rain characteristics such as rain shape, direction and intensity. Second, there are no public benchmarks for quantitative comparisons on real rain images, which makes the current evaluation less objective. The core challenge is that real world rain/clean image pairs cannot be captured at the same time. In this paper, we address the single image rain removal problem in two ways. First, we propose a semi-automatic method that incorporates temporal priors and human supervision to generate a high-quality clean image from each input sequence of real rain images. Using this method, we construct a large-scale dataset of $\sim$$29.5K$ rain/rain-free image pairs that covers a wide range of natural rain scenes. Second, to better cover the stochastic distribution of real rain streaks, we propose a novel SPatial Attentive Network (SPANet) to remove rain streaks in a local-to-global manner. Extensive experiments demonstrate that our network performs favorably against the state-of-the-art deraining methods.
Tasks	Rain Removal, Single Image Deraining
Published	2019-04-02
URL	https://arxiv.org/abs/1904.01538v2
PDF	https://arxiv.org/pdf/1904.01538v2.pdf
PWC	https://paperswithcode.com/paper/spatial-attentive-single-image-deraining-with
Repo	https://github.com/stevewongv/SPANet
Framework	pytorch

ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations


Title	ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations
Authors	Ksenia Korovina, Sailun Xu, Kirthevasan Kandasamy, Willie Neiswanger, Barnabas Poczos, Jeff Schneider, Eric P. Xing
Abstract	In applications such as molecule design or drug discovery, it is desirable to have an algorithm which recommends new candidate molecules based on the results of past tests. These molecules first need to be synthesized and then tested for objective properties. We describe ChemBO, a Bayesian optimization framework for generating and optimizing organic molecules for desired molecular properties. While most existing data-driven methods for this problem do not account for sample efficiency or fail to enforce realistic constraints on synthesizability, our approach explores the synthesis graph in a sample-efficient way and produces synthesizable candidates. We implement ChemBO as a Gaussian process model and explore existing molecular kernels for it. Moreover, we propose a novel optimal-transport based distance and kernel that accounts for graphical information explicitly. In our experiments, we demonstrate the efficacy of the proposed approach on several molecular optimization problems.
Tasks	Drug Discovery
Published	2019-08-05
URL	https://arxiv.org/abs/1908.01425v2
PDF	https://arxiv.org/pdf/1908.01425v2.pdf
PWC	https://paperswithcode.com/paper/chembo-bayesian-optimization-of-small-organic
Repo	https://github.com/cyclone923/ChemBo
Framework	none