janhuenermann/neurojs

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

4371 stars · 363 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

neurojs is a JavaScript deep learning framework for the browser with a focus on reinforcement learning, supporting deep Q-networks and actor-critic models with replay buffers. It includes demos such as a 2D self-driving car, but the project is explicitly no longer maintained.

Use cases

• train neural networks in the browser with javascript
• run reinforcement learning experiments like deep q-learning
• build a self-driving car simulation demo
• learn how actor-critic and DDPG algorithms work
• visualize neural network training in the browser

When to choose

• you need a small, readable codebase to study reinforcement learning internals
• you want browser-only neural network demos without server dependencies
• you're following the self-driving car tutorial

When to avoid

• you need a maintained production deep learning library - use TensorFlow.js instead
• you need GPU acceleration or large-scale training
• you need LSTM or modern architectures, which were never implemented

Facets

library · maturity abandoned

machine-learning deep-learning reinforcement-learning deep-learning machine-learning reinforcement-learning browser-extensions browser neural-networks deep-q-networks actor-critic self-driving-car browser-based-ml unmaintained javascript nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="janhuenermann/neurojs")