# MatthewJA/Inverse-Reinforcement-Learning

Implementations of selected inverse reinforcement learning algorithms.

Repository: https://github.com/MatthewJA/Inverse-Reinforcement-Learning
Canonical: https://ross.abutalabs.com/products/inverse-reinforcement-learning
Language: Python
License: MIT
License Family: permissive
Topics: inverse-reinforcement-learning, reinforcement-learning
Last push: 2022-10-21T07:40:38+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3968, "days_push": 1412, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1088, forks 236 (observed 2026-08-28T04:03:32.508316+00:00)

## What it is
A Python library implementing selected inverse reinforcement learning (IRL) algorithms, including linear programming IRL, maximum entropy IRL, and deep maximum entropy IRL. It also provides MDP domains such as Gridworld and Objectworld for experimentation.

## Use cases
- recover reward functions from expert trajectories
- run maximum entropy inverse reinforcement learning
- experiment with deep maxent IRL on MDPs
- learn how IRL algorithms work with reference implementations
- evaluate recovered rewards with expected value difference

## When to choose
- you need classic IRL algorithms (linear programming, maxent, deep maxent) in Python
- you want a small, citable research codebase for IRL experiments
- you need Gridworld/Objectworld testbeds for reward recovery

## When to avoid
- you need modern deep RL frameworks or GPU-accelerated training at scale
- you want actively maintained code with recent dependency support (Theano is deprecated)
- you need production-ready IRL tooling rather than research code

## Facets
- artifact type: library
- maturity: maintenance
- function: reinforcement-learning, machine-learning
- domain: reinforcement-learning, machine-learning
- platform: python
- tags: inverse-reinforcement-learning, maxent, gridworld, reward-recovery, mdp, algorithms

## Member repositories
- MatthewJA/Inverse-Reinforcement-Learning (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.508316+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:49:20.735086+00:00, confidence not recorded.
  - readme: https://github.com/MatthewJA/Inverse-Reinforcement-Learning (fetched 2026-08-28T04:03:32.508316+00:00, sha 35d2e8052098)
- Data as of 2026-08-30T08:39:29.467469+00:00.