A curated list of Decision Transformer resources (continually updated)

Code accompanying the paper Pretraining Language Models with Human Preferences

[ECCV2022] [T-PAMI] StARformer: Transformer with State-Action-Reward Representations.

Learning to Modulate pre-trained Models in RL (Decision Transformer, LoRA, Fine-tuning)

The official implementation of "Transformer in Transformer as Backbone for Deep Reinforcement Learning"

Implementation of Multi-Game Decision Transformers in PyTorch

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

Retrieval-Augmented Decision Transformer: External Memory for In-context RL

This repository is used to generate data and evaluate Decision Transformers on the CityLearn (Challenge 2022) environment for urban energy management

stable-baselines with JAX & Haiku

Code release for the ICLR 2023 conference paper "DeFog: Decision Transformer under Random Frame Dropping"

Reinforcement learning auto-bidding library for research and production.

Code for paper Sub-goal Distillation: A Method to Improve Small Language Agents, accepted at CoLLAs 2024.

Deep Reinforcement Learning AI to play the SuperTuxKart race game using a Decision Transformer

Applying regular transformer and decision transformer on solving the Rubik's cube. A paper is also written to document the results

ViAD - Visualize, Analyse and Decide methodology. It consists of leveraging Vision Transformers to get the raw visual perception of the environment, an LLM for comprehension and strategic planning, and Decision Transformers to implement low level execution leveraging memory-based learning.

Implementation of the decision tranformer paper in tinygrad

ATAG - Automated Test Automation Generation

Spiking Decision Transformer: Efficient Reinforcement Learning with Event-Based Sequence Modeling

Reinforcement Learning is often complex, but Decision Transformers frame offline RL as a sequence modeling problem. This repo provides the complete stack -- from development to deployment of Decision Transformers.

some experiments with training and fine-tuning decision transformer

Implementation of a Decision Transformer for power-grid control in L2RPN. The model learns from offline Tutor/Junior demonstrations and continues training online, serving as a drop-in replacement for PPO in the original pipeline proposed by aspirin96.

Implementing Decision Transformers from scratch for offline RL, benchmarking return-conditioned policies against Behavior Cloning.

Pilaf is a Backgammon agent using Decision Transformers and Offline RL