Search Results for author: Guiliang Liu

Found 10 papers, 3 papers with code

Benchmarking Constraint Inference in Inverse Reinforcement Learning

1 code implementation20 Jun 2022 Guiliang Liu, Yudong Luo, Ashish Gaurav, Kasra Rezaee, Pascal Poupart

When deploying Reinforcement Learning (RL) agents into a physical system, we must ensure that these agents are well aware of the underlying constraints.

Autonomous Driving reinforcement-learning

Learning Soft Constraints From Constrained Expert Demonstrations

no code implementations2 Jun 2022 Ashish Gaurav, Kasra Rezaee, Guiliang Liu, Pascal Poupart

We consider the setting where the reward function is given, and the constraints are unknown, and propose a method that is able to recover these constraints satisfactorily from the expert data.

Learning Tree Interpretation from Object Representation for Deep Reinforcement Learning

no code implementations NeurIPS 2021 Guiliang Liu, Xiangyu Sun, Oliver Schulte, Pascal Poupart

We propose a Represent And Mimic (RAMi) framework for training 1) an identifiable latent representation to capture the independent factors of variation for the objects and 2) a mimic tree that extracts the causal impact of the latent features on DRL action values.

reinforcement-learning

Distributional Reinforcement Learning with Monotonic Splines

no code implementations ICLR 2022 Yudong Luo, Guiliang Liu, Haonan Duan, Oliver Schulte, Pascal Poupart

Distributional Reinforcement Learning (RL) differs from traditional RL by estimating the distribution over returns to capture the intrinsic uncertainty of MDPs.

Distributional Reinforcement Learning reinforcement-learning

NTS-NOTEARS: Learning Nonparametric Temporal DAGs With Time-Series Data and Prior Knowledge

1 code implementation9 Sep 2021 Xiangyu Sun, Guiliang Liu, Pascal Poupart, Oliver Schulte

We propose a score-based DAG structure learning method for time-series data that captures linear, nonlinear, lagged and instantaneous relations among variables while ensuring acyclicity throughout the entire graph.

Time Series

Learning Agent Representations for Ice Hockey

no code implementations NeurIPS 2020 Guiliang Liu, Oliver Schulte, Pascal Poupart, Mike Rudd, Mehrsan Javan

This paper develops a new approach for agent representations, based on a Markov game model, that is tailored towards applications in professional ice hockey.

Sports Analytics

Cracking the Black Box: Distilling Deep Sports Analytics

1 code implementation4 Jun 2020 Xiangyu Sun, Jack Davis, Oliver Schulte, Guiliang Liu

This paper addresses the trade-off between Accuracy and Transparency for deep learning applied to sports analytics.

Sports Analytics

Toward Interpretable Deep Reinforcement Learning with Linear Model U-Trees

no code implementations16 Jul 2018 Guiliang Liu, Oliver Schulte, Wang Zhu, Qingcan Li

An LMUT is learned using a novel on-line algorithm that is well-suited for an active play setting, where the mimic learner observes an ongoing interaction between the neural net and the environment.

reinforcement-learning

Deep Reinforcement Learning in Ice Hockey for Context-Aware Player Evaluation

no code implementations26 May 2018 Guiliang Liu, Oliver Schulte

To assess a player's overall performance, we introduce a novel Game Impact Metric (GIM) that aggregates the values of the player's actions.

reinforcement-learning

Cannot find the paper you are looking for? You can Submit a new open access paper.