paper-with-me

홈 › Papers

Meta-DT: Offline Meta-RL as Conditional Sequence Modeling with World Model Disentanglement

2024-10-15 · Zhi Wang, Li Zhang, Wenhao Wu, Yuanheng Zhu, Dongbin Zhao, Chunlin Chen

A longstanding goal of artificial general intelligence is highly capable generalists that can learn from diverse experiences and generalize to unseen tasks. The language and vision communities have seen remarkable progress toward this trend by scaling up transformer-based models trained on massive datasets, while reinforcement learning (RL) agents still suffer from poor generalization capacity under such paradigms. To tackle this challenge, we propose Meta Decision Transformer (Meta-DT), which leverages the sequential modeling ability of the transformer architecture and robust task representation learning via world model disentanglement to achieve efficient generalization in offline meta-RL. We pretrain a context-aware world model to learn a compact task representation, and inject it as a contextual condition to the causal transformer to guide task-oriented sequence generation. Then, we subtly utilize history trajectories generated by the meta-policy as a self-guided prompt to exploit the architectural inductive bias. We select the trajectory segment that yields the largest prediction error on the pretrained world model to construct the prompt, aiming to encode task-specific information complementary to the world model maximally. Notably, the proposed framework eliminates the requirement of any expert demonstration or domain knowledge at test time. Experimental results on MuJoCo and Meta-World benchmarks across various dataset types show that Meta-DT exhibits superior few and zero-shot generalization capacity compared to strong baselines while being more practical with fewer prerequisites. Our code is available at https://github.com/NJU-RL/Meta-DT.

📄 PDF Abstract BibTeX arXiv:2410.11448

Code (1)

nju-rl/meta-dt 공식 구현

Tasks

DisentanglementInductive BiasMuJoCoReinforcement Learning (RL)Representation LearningZero-shot Generalization

Methods 이 논문이 사용한 방법론

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Residual Connection 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Adam 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

MetaDiffuser: Diffusion Model as Conditional Planner for Offline Meta-RL

2023-05-31 · Fei Ni, Jianye Hao, Yao Mu, Yifu Yuan 외

Recently, diffusion model shines as a promising backbone for the sequence modeling paradigm in offline reinforcement learning(RL). However, these works mostly lack the generalization ability across tasks with reward or d…

MuJoCoReinforcement Learning (RL)

Contextual Transformer for Offline Meta Reinforcement Learning

2022-11-15 · Runji Lin, Ye Li, Xidong Feng, Zhaowei Zhang 외

The pretrain-finetuning paradigm in large-scale sequence models has made significant progress in natural language processing and computer vision tasks. However, such a paradigm is still hindered by several challenges in …

D4RLMeta Reinforcement LearningMuJoCoOffline RL+5

Meta-Black-Box-Optimization through Offline Q-function Learning

2025-05-04 · Zeyuan Ma, Zhiguang Cao, Zhou Jiang, Hongshu Guo 외

Recent progress in Meta-Black-Box-Optimization (MetaBBO) has demonstrated that using RL to learn a meta-level policy for dynamic algorithm configuration (DAC) over an optimization task distribution could significantly en…

BenchmarkingMambaQ-Learning

Unsupervised Meta-Testing with Conditional Neural Processes for Hybrid Meta-Reinforcement Learning

2025-06-04 · Suzan Ece Ada, Emre Ugur

We introduce Unsupervised Meta-Testing with Conditional Neural Processes (UMCNP), a novel hybrid few-shot meta-reinforcement learning (meta-RL) method that uniquely combines, yet distinctly separates, parameterized polic…

continuous-controlContinuous ControlMeta Reinforcement Learning

Meta-World Conditional Neural Processes

2023-02-20 · Suzan Ece Ada, Emre Ugur

We propose Meta-World Conditional Neural Processes (MW-CNP), a conditional world model generator that leverages sample efficiency and scalability of Conditional Neural Processes to enable an agent to sample from its own …

Few-Shot LearningHallucination