# uber-research/deep-neuroevolution

Deep Neuroevolution

Repository: https://github.com/uber-research/deep-neuroevolution
Canonical: https://ross.abutalabs.com/products/deep-neuroevolution
Language: Python
License: NOASSERTION
License Family: other
Topics: deep-neuroevolution, ai, machine-learning, reinforcement-learning
Last push: 2024-01-08T03:03:13+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3164, "days_push": 968, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1667, forks 297 (observed 2026-08-28T04:05:19.649831+00:00)

## What it is
Distributed implementations of deep neuroevolution algorithms (ES, NS-ES, NSR-ES, DeepGA, Random Search) from Uber AI Labs research papers, for training deep neural networks on reinforcement learning tasks. It includes a VINE visualization tool for inspecting neuroevolution experiments and a GPU-accelerated implementation.

## Use cases
- train deep neural networks for reinforcement learning with evolution strategies
- run genetic algorithms on Atari games
- compare neuroevolution against gradient-based RL methods
- visualize neuroevolution experiment populations with VINE
- run distributed RL experiments on AWS with Redis
- reproduce Uber AI deep neuroevolution paper results

## When to choose
- you want to experiment with evolution strategies or genetic algorithms for deep RL
- you need to reproduce or extend the Deep Neuroevolution papers
- you want population-based RL training that avoids gradient backpropagation
- you need an interactive visualization of neuroevolution search

## When to avoid
- you need standard gradient-based RL like PPO or DQN
- you need a maintained production library with active support
- you cannot obtain a MuJoCo license for humanoid experiments
- you need Windows support or a simple pip-installable package

## Facets
- artifact type: library
- maturity: maintenance
- function: reinforcement-learning, machine-learning, data-visualization, gpu-computing
- domain: reinforcement-learning, machine-learning, artificial-intelligence
- platform: python, cloud
- tags: neuroevolution, evolution-strategies, genetic-algorithms, atari, mujoco, research-code, distributed-computing, redis, research, linux, macos, docker

## Member repositories
- uber-research/deep-neuroevolution (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:19.649831+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 2026-08-30T03:42:54.192551+00:00, confidence not recorded.
  - readme: https://github.com/uber-research/deep-neuroevolution (fetched 2026-08-28T04:05:19.649831+00:00, sha 336724f1045b)
- Data as of 2026-08-30T08:39:29.467469+00:00.