Procedural Content Generation: Better Benchmarks for Transfer Reinforcement Learning

被引:3
|
作者
Muller-Brockhausen, Matthias [1 ]
Preuss, Mike [1 ]
Plaat, Aske [1 ]
机构
[1] Leiden Univ, Leiden Inst Adv Comp Sci, Leiden, Netherlands
关键词
Transfer; Reinforcement Learning; Benchmarks; Procedural Content Generation; FRAMEWORK; AI;
D O I
10.1109/COG52621.2021.9619000
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The idea of transfer in reinforcement learning (TRL) is intriguing: being able to transfer knowledge from one problem to another problem without learning everything from scratch. This promises quicker learning and learning more complex methods. To gain an insight into the field and to detect emerging trends, we performed a database search. We note a surprisingly late adoption of deep learning that starts in 2018. The introduction of deep learning has not yet solved the greatest challenge of TRL: generalization. Transfer between different domains works well when domains have strong similarities (e.g. MountainCar to Cartpole), and most TRL publications focus on different tasks within the same domain that have few differences. Most TRL applications we encountered compare their improvements against self-defined baselines, and the field is still missing unified benchmarks. We consider this to be a disappointing situation. For the future, we note that: (1) A clear measure of task similarity is needed. (2) Generalization needs to improve. Promising approaches merge deep learning with planning via MCTS or introduce memory through LSTMs. (3) The lack of benchmarking tools will be remedied to enable meaningful comparison and measure progress. Already Alchemy and Meta-World are emerging as interesting benchmark suites. We note that another development, the increase in procedural content generation (PCG), can improve both benchmarking and generalization in TRL.
引用
收藏
页码:924 / 931
页数:8
相关论文
共 50 条
  • [21] Continuous Procedural Network of Roads Generation using L-Systems and Reinforcement Learning
    Paduraru, Ciprian
    Paduraru, Miruna
    Iordache, Stefan
    PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON SOFTWARE TECHNOLOGIES (ICSOFT), 2022, : 425 - 432
  • [22] An Adjectival Interface for Procedural Content Generation
    Hultquist, Carl
    Gain, James
    Cairns, David
    INTELLIGENT COMPUTER GRAPHICS 2009, 2009, 240 : 143 - +
  • [23] Capturing Local and Global Patterns in Procedural Content Generation via Machine Learning
    Volz, Vanessa
    Justesen, Niels
    Snodgrass, Sam
    Asadi, Sahar
    Purmonen, Sami
    Holmgard, Christoffer
    Togelius, Julian
    Risi, Sebastian
    2020 IEEE CONFERENCE ON GAMES (IEEE COG 2020), 2020, : 399 - 406
  • [24] Procedural Content Generation for Games: A Survey
    Hendrikx, Mark
    Meijer, Sebastiaan
    van der Velden, Joeri
    Iosup, Alexandru
    ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS, 2013, 9 (01)
  • [25] RL Unplugged: A Suite of Benchmarks for Offline Reinforcement Learning
    Gulcehre, Caglar
    Wang, Ziyu
    Novikov, Alexander
    Le Paine, Tom
    Colmenarejo, Sergio Gomez
    Zolna, Konrad
    Agarwal, Rishabh
    Merel, Josh
    Mankowitz, Daniel
    Paduraru, Cosmin
    Dulac-Arnold, Gabriel
    Li, Jerry
    Norouzi, Mohammad
    Hoffman, Matt
    Heess, Nicolas
    de Freitas, Nando
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [26] Sub-Resolution Assist Feature Generation with Reinforcement Learning and Transfer Learning
    Liu, Guan-Ting
    Tai, Wei-Chen
    Lin, Yi-Ting
    Jiang, Iris Hui-Ru
    Shiely, James P.
    Cheng, Pu-Jen
    2022 IEEE/ACM INTERNATIONAL CONFERENCE ON COMPUTER AIDED DESIGN, ICCAD, 2022,
  • [27] Emotional editing constraint conversation content generation based on reinforcement learning
    Sun, Xiao
    Li, Jia
    Wei, Xing
    Li, Changliang
    Tao, Jianhua
    INFORMATION FUSION, 2020, 56 (56) : 70 - 80
  • [28] Envisioning better benchmarks for machine learning PDE solvers
    Brandstetter, Johannes
    NATURE MACHINE INTELLIGENCE, 2025, 7 (01) : 2 - 3
  • [29] Procedural Content Generation via Machine Learning in 2D Indoor Scene
    Jezek, Bruno
    Ouhrabka, Adam
    Slaby, Antonin
    AUGMENTED REALITY, VIRTUAL REALITY, AND COMPUTER GRAPHICS, AVR 2020, PT I, 2020, 12242 : 34 - 49
  • [30] Guest Editorial: Procedural Content Generation in Games
    Togelius, Julian
    Whitehead, Jim
    Bidarra, Rafael
    IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES, 2011, 3 (03) : 169 - 171