Learning Pareto Set for Multi-Objective Continuous Robot Control

1 code implementation • 27 Jun 2024 • Tianye Shu, Ke Shang, Cheng Gong, Yang Nan, Hisao Ishibuchi

For a control problem with multiple conflicting objectives, there exists a set of Pareto-optimal policies called the Pareto set instead of a single optimal policy.

Multi-Objective Reinforcement Learning

State Space Closure: Revisiting Endless Online Level Generation via Reinforcement Learning

1 code implementation • 6 Dec 2022 • Ziqi Wang, Tianye Shu, Jialin Liu

Concluding our outcomes and analysis, future work on endless online level generation via reinforcement learning should address the issue of diversity while assuring the occurrence of state space closure and quality.

Diversity reinforcement-learning +2

Effects of Archive Size on Computation Time and Solution Quality for Multi-Objective Optimization

1 code implementation • 7 Sep 2022 • Tianye Shu, Ke Shang, Hisao Ishibuchi, Yang Nan

In this study, we examine the effects of the archive size on three aspects: (i) the quality of the selected final solution set, (ii) the total computation time for the archive maintenance and the final solution set selection, and (iii) the required memory size.

Learning to Approximate: Auto Direction Vector Set Generation for Hypervolume Contribution Approximation

1 code implementation • 18 Jan 2022 • Ke Shang, Tianye Shu, Hisao Ishibuchi

The learned direction vector set can then be used in the $R_2^{\text{HVC}}$ indicator to improve its approximation quality.

Benchmarking Subset Selection from Large Candidate Solution Sets in Evolutionary Multi-objective Optimization

2 code implementations • 18 Jan 2022 • Ke Shang, Tianye Shu, Hisao Ishibuchi, Yang Nan, Lie Meng Pang

This paper aims to fill this research gap by proposing a benchmark test suite for subset selection from large candidate solution sets, and comparing some representative methods using the proposed test suite.

Benchmarking

Experience-Driven PCG via Reinforcement Learning: A Super Mario Bros Study

1 code implementation • 30 Jun 2021 • Tianye Shu, Jialin Liu, Georgios N. Yannakakis

In particular, the RL designers of Super Mario Bros generate and concatenate level segments while considering the diversity among the segments.

Diversity reinforcement-learning +1

Reinforcement Learning with Dual-Observation for General Video Game Playing

1 code implementation • 11 Nov 2020 • Chengpeng Hu, Ziqi Wang, Tianye Shu, Hao Tong, Julian Togelius, Xin Yao, Jialin Liu

Our proposed technique is implemented with three state-of-the-art reinforcement learning algorithms and tested on the game set of the 2020 General Video Game AI Learning Competition.

Decision Making reinforcement-learning +2

A Novel CNet-assisted Evolutionary Level Repairer and Its Applications to Super Mario Bros

4 code implementations • 13 May 2020 • Tianye Shu, Ziqi Wang, Jialin Liu, Xin Yao

However, defective levels with illegal patterns may be generated due to the violation of constraints for level design.