Learning to Modulate pre-trained Models in RL

被引:0
|
作者
Schmied, Thomas [1 ,2 ]
Hofmarcher, Markus [3 ]
Paischer, Fabian [1 ,2 ]
Pascanu, Razvan [4 ,5 ]
Hochreiter, Sepp [1 ,2 ]
机构
[1] ELLIS Unit Linz, Linz, Austria
[2] Inst Machine Learning, LIT AI Lab, Linz, Austria
[3] Johannes Kepler Univ Linz, JKU LIT SAL eSPML Lab, Inst Machine Learning, Linz, Austria
[4] Google DeepMind, London, England
[5] UCL, London, England
基金
欧盟地平线“2020”;
关键词
NEURAL-NETWORKS; REINFORCEMENT;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Reinforcement Learning (RL) has been successful in various domains like robotics, game playing, and simulation. While RL agents have shown impressive capabilities in their specific tasks, they insufficiently adapt to new tasks. In supervised learning, this adaptation problem is addressed by large-scale pre-training followed by fine-tuning to new down-stream tasks. Recently, pre-training on multiple tasks has been gaining traction in RL. However, fine-tuning a pre-trained model often suffers from catastrophic forgetting. That is, the performance on the pre-training tasks deteriorates when fine-tuning on new tasks. To investigate the catastrophic forgetting phenomenon, we first jointly pre-train a model on datasets from two benchmark suites, namely Meta-World and DMControl. Then, we evaluate and compare a variety of fine-tuning methods prevalent in natural language processing, both in terms of performance on new tasks, and how well performance on pre-training tasks is retained. Our study shows that with most fine-tuning approaches, the performance on pre-training tasks deteriorates significantly. Therefore, we propose a novel method, Learning-to-Modulate (L2M), that avoids the degradation of learned skills by modulating the information flow of the frozen pre-trained model via a learnable modulation pool. Our method achieves state-of-the-art performance on the Continual-World benchmark, while retaining performance on the pre-training tasks. Finally, to aid future research in this area, we release a dataset encompassing 50 Meta-World and 16 DMControl tasks.
引用
收藏
页数:35
相关论文
共 50 条
  • [31] Generalization of vision pre-trained models for histopathology
    Milad Sikaroudi
    Maryam Hosseini
    Ricardo Gonzalez
    Shahryar Rahnamayan
    H. R. Tizhoosh
    [J]. Scientific Reports, 13
  • [32] Pre-trained models: Past, present and future
    Han, Xu
    Zhang, Zhengyan
    Ding, Ning
    Gu, Yuxian
    Liu, Xiao
    Huo, Yuqi
    Qiu, Jiezhong
    Yao, Yuan
    Zhang, Ao
    Zhang, Liang
    Han, Wentao
    Huang, Minlie
    Jin, Qin
    Lan, Yanyan
    Liu, Yang
    Liu, Zhiyuan
    Lu, Zhiwu
    Qiu, Xipeng
    Song, Ruihua
    Tang, Jie
    Wen, Ji-Rong
    Yuan, Jinhui
    Zhao, Wayne Xin
    Zhu, Jun
    [J]. AI OPEN, 2021, 2 : 225 - 250
  • [33] Interpreting Art by Leveraging Pre-Trained Models
    Penzel, Niklas
    Denzler, Joachim
    [J]. 2023 18TH INTERNATIONAL CONFERENCE ON MACHINE VISION AND APPLICATIONS, MVA, 2023,
  • [34] Lottery Jackpots Exist in Pre-Trained Models
    Zhang, Yuxin
    Lin, Mingbao
    Zhong, Yunshan
    Chao, Fei
    Ji, Rongrong
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (12) : 14990 - 15004
  • [35] LaoPLM: Pre-trained Language Models for Lao
    Lin, Nankai
    Fu, Yingwen
    Yang, Ziyu
    Chen, Chuwei
    Jiang, Shengyi
    [J]. LREC 2022: THIRTEEN INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION, 2022, : 6506 - 6512
  • [36] Generalization of vision pre-trained models for histopathology
    Sikaroudi, Milad
    Hosseini, Maryam
    Gonzalez, Ricardo
    Rahnamayan, Shahryar
    Tizhoosh, H. R.
    [J]. SCIENTIFIC REPORTS, 2023, 13 (01)
  • [37] PhoBERT: Pre-trained language models for Vietnamese
    Dat Quoc Nguyen
    Anh Tuan Nguyen
    [J]. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, EMNLP 2020, 2020, : 1037 - 1042
  • [38] HinPLMs: Pre-trained Language Models for Hindi
    Huang, Xixuan
    Lin, Nankai
    Li, Kexin
    Wang, Lianxi
    Gan, Suifu
    [J]. 2021 INTERNATIONAL CONFERENCE ON ASIAN LANGUAGE PROCESSING (IALP), 2021, : 241 - 246
  • [39] Semantic Programming by Example with Pre-trained Models
    Verbruggen, Gust
    Le, Vu
    Gulwani, Sumit
    [J]. PROCEEDINGS OF THE ACM ON PROGRAMMING LANGUAGES-PACMPL, 2021, 5 (OOPSLA):
  • [40] Evaluating Commonsense in Pre-Trained Language Models
    Zhou, Xuhui
    Zhang, Yue
    Cui, Leyang
    Huang, Dandan
    [J]. THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 9733 - 9740