Cross-Episodic Curriculum for Transformer Agents

• URL: http://arxiv.org/abs/2310.08549v1
• Date: Thu, 12 Oct 2023 17:45:05 GMT
• Title: Cross-Episodic Curriculum for Transformer Agents
• Authors: Lucy Xiaoyang Shi and Yunfan Jiang and Jake Grigsby and Linxi "Jim" Fan and Yuke Zhu
• Abstract summary: We present a new algorithm, Cross-Episodic Curriculum ( CEC), to boost the learning efficiency and generalization of Transformer agents. Central to CEC is the placement of cross-episodic experiences into a Transformer's context. CEC constructs curricula that encapsulate learning progression and proficiency increase across episodes.
• Score: 26.240903251696874
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present a new algorithm, Cross-Episodic Curriculum (CEC), to boost the learning efficiency and generalization of Transformer agents. Central to CEC is the placement of cross-episodic experiences into a Transformer's context, which forms the basis of a curriculum. By sequentially structuring online learning trials and mixed-quality demonstrations, CEC constructs curricula that encapsulate learning progression and proficiency increase across episodes. Such synergy combined with the potent pattern recognition capabilities of Transformer models delivers a powerful cross-episodic attention mechanism. The effectiveness of CEC is demonstrated under two representative scenarios: one involving multi-task reinforcement learning with discrete control, such as in DeepMind Lab, where the curriculum captures the learning progression in both individual and progressively complex settings; and the other involving imitation learning with mixed-quality data for continuous control, as seen in RoboMimic, where the curriculum captures the improvement in demonstrators' expertise. In all instances, policies resulting from CEC exhibit superior performance and strong generalization. Code is open-sourced at https://cec-agent.github.io/ to facilitate research on Transformer agent learning.

Related papers

• CNN-Transformer Rectified Collaborative Learning for Medical Image Segmentation [60.08541107831459]
  This paper proposes a CNN-Transformer rectified collaborative learning framework to learn stronger CNN-based and Transformer-based models for medical image segmentation. Specifically, we propose a rectified logit-wise collaborative learning (RLCL) strategy which introduces the ground truth to adaptively select and rectify the wrong regions in student soft labels. We also propose a class-aware feature-wise collaborative learning (CFCL) strategy to achieve effective knowledge transfer between CNN-based and Transformer-based models in the feature space.
  arXiv  Detail & Related papers  (2024-08-25T01:27:35Z)
• Variational Curriculum Reinforcement Learning for Unsupervised Discovery of Skills [25.326624139426514]
  We propose a novel approach to unsupervised skill discovery based on information theory, called Value Uncertainty Vari Curriculum Curriculum (VUVC) We prove that, under regularity conditions, VUVC accelerates the increase of entropy in the visited states compared to the uniform curriculum. We also demonstrate that the skills discovered by our method successfully complete a real-world robot navigation task in a zero-shot setup.
  arXiv  Detail & Related papers  (2023-10-30T10:34:25Z)
• How Do Transformers Learn In-Context Beyond Simple Functions? A Case Study on Learning with Representations [98.7450564309923]
  This paper takes initial steps on understanding in-context learning (ICL) in more complex scenarios, by studying learning with representations. We construct synthetic in-context learning problems with a compositional structure, where the label depends on the input through a possibly complex but fixed representation function. We show theoretically the existence of transformers that approximately implement such algorithms with mild depth and size.
  arXiv  Detail & Related papers  (2023-10-16T17:40:49Z)
• Integrating Contrastive Learning into a Multitask Transformer Model for Effective Domain Adaptation [4.157415305926585]
  We propose a novel domain adaptation technique that embodies a multitask framework with SER as the primary task. We show that our proposed model achieves state-of-the-art performance in SER within cross-corpus scenarios.
  arXiv  Detail & Related papers  (2023-10-07T06:41:29Z)
• Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection [88.23337313766353]
  This work first provides a comprehensive statistical theory for transformers to perform ICL. We show that transformers can implement a broad class of standard machine learning algorithms in context. A emphsingle transformer can adaptively select different base ICL algorithms.
  arXiv  Detail & Related papers  (2023-06-07T17:59:31Z)
• CoopInit: Initializing Generative Adversarial Networks via Cooperative Learning [50.90384817689249]
  CoopInit is a cooperative learning-based strategy that can quickly learn a good starting point for GANs. We demonstrate the effectiveness of the proposed approach on image generation and one-sided unpaired image-to-image translation tasks.
  arXiv  Detail & Related papers  (2023-03-21T07:49:32Z)
• Weakly Supervised Disentangled Representation for Goal-conditioned Reinforcement Learning [15.698612710580447]
  We propose a skill learning framework DR-GRL that aims to improve the sample efficiency and policy generalization. In a weakly supervised manner, we propose a Spatial Transform AutoEncoder (STAE) to learn an interpretable and controllable representation. We empirically demonstrate that DR-GRL significantly outperforms the previous methods in sample efficiency and policy generalization.
  arXiv  Detail & Related papers  (2022-02-28T09:05:14Z)
• Learning Multi-Objective Curricula for Deep Reinforcement Learning [55.27879754113767]
  Various automatic curriculum learning (ACL) methods have been proposed to improve the sample efficiency and final performance of deep reinforcement learning (DRL) In this paper, we propose a unified automatic curriculum learning framework to create multi-objective but coherent curricula. In addition to existing hand-designed curricula paradigms, we further design a flexible memory mechanism to learn an abstract curriculum.
  arXiv  Detail & Related papers  (2021-10-06T19:30:25Z)
• 3D Human Action Representation Learning via Cross-View Consistency Pursuit [52.19199260960558]
  We propose a Cross-view Contrastive Learning framework for unsupervised 3D skeleton-based action Representation (CrosSCLR) CrosSCLR consists of both single-view contrastive learning (SkeletonCLR) and cross-view consistent knowledge mining (CVC-KM) modules, integrated in a collaborative learning manner.
  arXiv  Detail & Related papers  (2021-04-29T16:29:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.