AnyMorph: Learning Transferable Polices By Inferring Agent Morphology

被引:0
作者
Trabucco, Brandon [1 ,2 ]
Phielipp, Mariano [2 ]
Berseth, Glen [3 ]
机构
[1] Carnegie Mellon Univ, Machine Learning Dept, Pittsburgh, PA 15213 USA
[2] Intel AI, San Diego, CA USA
[3] Mila, Montreal, PQ, Canada
来源
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162 | 2022年
关键词
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
The prototypical approach to reinforcement learning involves training policies tailored to a particular agent from scratch for every new morphology. Recent work aims to eliminate the re-training of policies by investigating whether a morphology-agnostic policy, trained on a diverse set of agents with similar task objectives, can be transferred to new agents with unseen morphologies without re-training. This is a challenging problem that required previous approaches to use hand-designed descriptions of the new agent's morphology. Instead of hand-designing this description, we propose a data-driven method that learns a representation of morphology directly from the reinforcement learning objective. Ours is the first reinforcement learning algorithm that can train a policy to generalize to new agent morphologies without requiring a description of the agent's morphology in advance. We evaluate our approach on the standard benchmark for agent-agnostic control, and improve over the current state of the art in zero-shot generalization to new agents. Importantly, our method attains good performance without an explicit description of morphology.
引用
收藏
页数:15
相关论文
共 50 条
  • [41] Inferring Robot Morphology from Observation of Unscripted Movement
    Bell, Neil
    Seipp, Brian
    Oates, J. Tim
    Matuszek, Cynthia
    2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2019, : 9544 - 9551
  • [42] Species Delimitation: Inferring Gaps in Morphology across Geography
    Zapata, Felipe
    Jimenez, Ivan
    SYSTEMATIC BIOLOGY, 2012, 61 (02) : 179 - 194
  • [43] Inferring Commitment Semantics in Multi-Agent Interactions
    Chocron, Paula
    Schorlemmer, Marco
    PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS (AAMAS' 18), 2018, : 1150 - 1158
  • [44] Learning Transferable Architectures for Scalable Image Recognition
    Zoph, Barret
    Vasudevan, Vijay
    Shlens, Jonathon
    Le, Quoc V.
    2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 8697 - 8710
  • [45] Learning explicitly transferable representations for domain adaptation
    Jing, Mengmeng
    Li, Jingjing
    Lu, Ke
    Zhu, Lei
    Yang, Yang
    NEURAL NETWORKS, 2020, 130 : 39 - 48
  • [46] Learning Fair and Transferable Representations with Theoretical Guarantees
    Oneto, Luca
    Donini, Michele
    Pontil, Massimiliano
    Maurer, Andreas
    2020 IEEE 7TH INTERNATIONAL CONFERENCE ON DATA SCIENCE AND ADVANCED ANALYTICS (DSAA 2020), 2020, : 30 - 39
  • [47] GLoMo: Unsupervised Learning of Transferable Relational Graphs
    Yang, Zhilin
    Zhao, Jake
    Dhingra, Bhuwan
    He, Kaiming
    Cohen, William W.
    Salakhutdinov, Ruslan
    LeCun, Yann
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [48] An accurate and transferable machine learning potential for carbon
    Rowe, Patrick
    Deringer, Volker L.
    Gasparotto, Piero
    Csanyi, Gabor
    Michaelides, Angelos
    JOURNAL OF CHEMICAL PHYSICS, 2020, 153 (03)
  • [49] Learning Transferable Parameters for Unsupervised Domain Adaptation
    Han, Zhongyi
    Sun, Haoliang
    Yin, Yilong
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 31 : 6424 - 6439
  • [50] Learning Transferable Subspace for Human Motion Segmentation
    Wang, Lichen
    Ding, Zhengming
    Fu, Yun
    THIRTY-SECOND AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTIETH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / EIGHTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2018, : 4195 - 4202