博客 | 过去10年NIPS顶会强化学习论文（100多篇）汇总（2008-2018年）

“

本文原载于微信公众号：深度强化学习算法 ，AI研习社经授权转载。欢迎关注 深度学习强化算法 微信公众号、及 AI研习社博客专栏。

NIPS（NeurIPS），全称神经信息处理系统大会(Conference and Workshop on Neural Information Processing Systems)，是一个关于机器学习和计算神经科学的国际会议。该会议固定在每年的12月举行,由NIPS基金会主办。NIPS是机器学习领域的顶级会议。在中国计算机学会的国际学术会议排名中，NIPS为人工智能领域的A类会议,自1987年到2000年，NIPS都在美国的丹佛举办。而在2001年到2010年，NIPS的举办地则是在加拿大的温哥华。此后，NIPS分别在在西班牙的格兰纳达（2011年），太浩湖（Lake Tahoe）（2012年到2013年），以及加拿大的蒙特利尔（2014到2015年）

自从数年前深度学习流行以来，NIPS 成为学术界、产业界重点关注的学术会议之一，参会人数从 5 年前的 2000 人一度飙升到 2017 年的 8000 多人。除参会人员，2017 年 NIPS 的论文投稿也创造了历史新高，达到了 3240 篇。最近的统计显示，NIPS 2018 论文投稿数量高达 4900 篇，比去年又多了 1600 多篇，在过去几年中，各个领域文章特别多，本文汇总了过去10年NIPS会议接收的108篇强化学习领域的文章内容，具体总结如下：

从表中我们可以看出强化学习在2012年是一个分水岭，经历了火热之后开始衰退，然后从2015年开始一路攀升，达到了录取38篇的数量，下面是历届accept paper的题目list

2008年（3）

• Near-optimal Regret Bounds for Reinforcement Learning
• Stress, noradrenaline, and realistic prediction of mouse behaviour using reinforcement learning
• Optimization on a Budget: A Reinforcement Learning Approach

2009（3）

• Manifold Embeddings for Model-Based Reinforcement Learning under Partial Observability
• Skill Discovery in Continuous Reinforcement Learning Domains using Skill Chaining
• Training Factor Graphs with Reinforcement Learning for Efficient MAP Inference

2010年（5）

• Nonparametric Bayesian Policy Priors for Reinforcement Learning
• Constructing Skill Trees for Reinforcement Learning Agents from Demonstration Trajectories
• Feature Construction for Inverse Reinforcement Learning
• PAC-Bayesian Model Selection for Reinforcement Learning
• Interval Estimation for Reinforcement-Learning Algorithms in Continuous-State Domains

2011年（7）

• Nonlinear Inverse Reinforcement Learning with Gaussian Processes
• A Reinforcement Learning Theory for Homeostatic Regulation
• Action-Gap Phenomenon in Reinforcement Learning
• Optimal Reinforcement Learning for Gaussian Systems
• Reinforcement Learning using Kernel-Based Stochastic Factorization
• MAP Inference for Bayesian Inverse Reinforcement Learning
• Selecting the State-Representation in Reinforcement Learning

2012年(11)

• Bayesian Hierarchical Reinforcement Learning
• Exploration in Model-based Reinforcement Learning by Empirically Estimating Learning Progress
• Nonparametric Bayesian Inverse Reinforcement Learning for Multiple Reward Functions
• Inverse Reinforcement Learning through Structured Classification
• Efficient Bayes-Adaptive Reinforcement Learning using Sample-Based Search
• On-line Reinforcement Learning Using Incremental Kernel-Based Stochastic Factorization
• Online Regret Bounds for Undiscounted Continuous Reinforcement Learning
• Neurally Plausible Reinforcement Learning of Working Memory Tasks
• Transferring Expectations in Model-based Reinforcement Learning
• Efficient Reinforcement Learning for High Dimensional Linear Quadratic Systems
• Cost-Sensitive Exploration in Bayesian Reinforcement Learning

2013年(3)

• Reinforcement Learning in Robust Markov Decision Processes
• Policy Shaping: Integrating Human Feedback with Reinforcement Learning
• (More) Efficient Reinforcement Learning via Posterior Sampling

2014年(5)

• Model-based Reinforcement Learning and the Eluder Dimension
• Sparse Multi-Task Reinforcement Learning
• Difference of Convex Functions Programming for Reinforcement Learning
• Near-optimal Reinforcement Learning in Factored MDPs
• RAAM: The Benefits of Robustness in Approximating Aggregated MDPs in Reinforcement Learning

2015年(2)

• Sample Complexity of Episodic Fixed-Horizon Reinforcement Learning
• Inverse Reinforcement Learning with Locally Consistent Reward Functions

2016年(7)

• Tree-Structured Reinforcement Learning for Sequential Object Localization
• Safe and Efficient Off-Policy Reinforcement Learning
• Contextual-MDPs for PAC Reinforcement Learning with Rich Observations
• Learning to Communicate with Deep Multi-Agent Reinforcement Learning
• Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
• Cooperative Inverse Reinforcement Learning
• Linear Feature Encoding for Reinforcement Learning

2017年(24)

• Hybrid Reward Architecture for Reinforcement Learning
• Shallow Updates for Deep Reinforcement Learning
• Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning
• Optimistic posterior sampling for reinforcement learning: worst-case regret bounds
• Cold-Start Reinforcement Learning with Softmax Policy Gradient
• Log-normality and Skewness of Estimated State/Action Values in Reinforcement Learning
• Safe Model-based Reinforcement Learning with Stability Guarantees
• Data-Efficient Reinforcement Learning in Continuous State-Action Gaussian-POMDPs
• Deep Reinforcement Learning from Human Preferences
• EX2: Exploration with Exemplar Models for Deep Reinforcement Learning
• Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
• Compatible Reward Inverse Reinforcement Learning
• Bridging the Gap Between Value and Policy Based Reinforcement Learning
• Compatible Reward Inverse Reinforcement Learning
• Online Reinforcement Learning in Stochastic Games
• Reinforcement Learning under Model Mismatch
• A multi-agent reinforcement learning model of common-pool resource appropriation
• Imagination-Augmented Agents for Deep Reinforcement Learning
• Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
• Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning
• Repeated Inverse Reinforcement Learning
• A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning
• Dynamic Safe Interruptibility for Decentralized Multi-Agent Reinforcement Learning
• Agent Reinforcement Learning

2018年(38)

• The Importance of Sampling inMeta-Reinforcement Learning
• Learning Temporal Point Processes via Reinforcement Learning
• Data-Efficient Hierarchical Reinforcement Learning
• Fast deep reinforcement learning using online adjustments from the past
• Inference Aided Reinforcement Learning for Incentive Mechanism Design in Crowdsourcing
• A Lyapunov-based Approach to Safe Reinforcement Learning
• Reinforcement Learning of Theorem Proving
• Simple random search of static linear policies is competitive for reinforcement learning
• Meta-Gradient Reinforcement Learning
• Reinforcement Learning for Solving the Vehicle Routing Problem
• Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning
• REFUEL: Exploring Sparse Features in Deep Reinforcement Learning for Fast Disease Diagnosis
• Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents
• Distributed Multitask Reinforcement Learning with Quadratic Convergence
• Constrained Cross-Entropy Method for Safe Reinforcement Learning
• Reinforcement Learning with Multiple Experts: A Bayesian Model Combination Approach
• Verifiable Reinforcement Learning via Policy Extraction
• Deep Reinforcement Learning of Marked Temporal Point Processes
• Evolution-Guided Policy Gradient in Reinforcement Learning
• Meta-Reinforcement Learning of Structured Exploration Strategies
• Diversity-Driven Exploration Strategy for Deep Reinforcement Learning
• Genetic-Gated Networks for Deep Reinforcement Learning
• Visual Reinforcement Learning with Imagined Goals
• Unsupervised Video Object Segmentation for Deep Reinforcement Learning
• Total stochastic gradient algorithms and applications in reinforcement learning
• Fighting Boredom in Recommender Systems with Linear Reinforcement Learning
• Randomized Prior Functions for Deep Reinforcement Learning
• Scalable Coordinated Exploration in Concurrent Reinforcement Learning
• Multi-Agent Reinforcement Learning via Double Averaging Primal-Dual Optimization
• Negotiable Reinforcement Learning for Pareto Optimal Sequential Decision-Making
• Teaching Inverse Reinforcement Learners via Features and Demonstrations
• Hierarchical Reinforcement Learning for Zero-shot Generalization with Subtask Dependencies
• Lifelong Inverse Reinforcement Learning
• Multiple-Step Greedy Policies in Approximate and Online Reinforcement Learning
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
• Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion