Personal notes about scientific and research works on "Decision-Making for Autonomous Driving"

Simple (but often Strong) Baselines for POMDPs in PyTorch, ICML 2022

A framework to build and solve POMDP problems. Documentation: https://h2r.github.io/pomdp-py/

The DESPOT online POMDP solver

Partially Observable Process Gym

Clean baseline implementation of PPO using an episodic TransformerXL memory

A Modern Probabilistic Model Checker

Baseline implementation of recurrent PPO using truncated BPTT

Adaptive Attention Span for Reinforcement Learning

✍️ Python implementation of POMDP framework and PBVI & POMCP algorithms.

Challenging Memory-based Deep Reinforcement Learning Agents

Solving POMDP using Recurrent networks

Efficient Point-Based POMDP Planning by Approximating

When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment, NeurIPS 2023 (oral)

Online Replanning in Belief Space for Partially Observable Task and Motion Problems

Modeling agents with probabilistic programs

Official Implementation for the paper "SR-AIF: Solving Sparse-Reward Robotic Tasks from Pixels with Active Inference and World Models"

🎓 Adaptive AI study agent with POMDP belief state — OPEAA loop, Q-learning + LinUCB bandit policies, SM-2 spaced repetition, concept DAG. Streamlit web app + Chrome extension (MV3). Claude & free HF backends.

🔥 [ICLR 2026] Benchmark for robotic tabletop manipulation memory-intensive tasks

Rich literature review and discussion on the implementation of "Hierarchical Decision-Making for Autonomous Driving"