# lexfridman/deeptraffic

DeepTraffic is a deep reinforcement learning competition, part of the MIT Deep Learning series.

Repository: https://github.com/lexfridman/deeptraffic
Canonical: https://ross.abutalabs.com/products/deeptraffic
Homepage: https://selfdrivingcars.mit.edu/deeptraffic
Language: JavaScript
License: MIT
License Family: permissive
Topics: deep-learning, machine-learning, deep-reinforcement-learning, mit, self-driving-cars, deep-rl, tensorflow, convnetjs
Last push: 2023-08-01T23:09:33+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": 3162, "days_push": 1128, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1793, forks 278 (observed 2026-08-28T04:05:37.257083+00:00)

## What it is
DeepTraffic is a deep reinforcement learning competition from the MIT Deep Learning series where participants train neural networks to drive vehicles fast through dense highway traffic. The repository provides starter code snippets and agents to submit to the browser-based competition and leaderboard.

## Use cases
- learn deep reinforcement learning hands-on
- train a neural network agent for highway driving
- compete on a deep RL leaderboard
- teach a course on deep learning and autonomous vehicles
- experiment with hyperparameter tuning for RL
- simulate multi-agent dense traffic navigation

## When to choose
- you want an accessible browser-based deep RL competition
- you are teaching or learning reinforcement learning with a concrete autonomous driving task
- you want to benchmark RL agents on a shared traffic simulation

## When to avoid
- you need a production-grade autonomous driving stack
- you require a general-purpose RL training framework with modern tooling
- you need actively maintained code with recent updates

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: reinforcement-learning, machine-learning, deep-learning, simulation
- domain: reinforcement-learning, autonomous-vehicles, education, machine-learning, simulation
- platform: browser
- tags: deep-rl, self-driving-cars, competition, tensorflow, convnetjs, mit, traffic-simulation, javascript, web

## Member repositories
- lexfridman/deeptraffic (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:37.257083+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 2026-08-30T03:22:56.473002+00:00, confidence not recorded.
  - readme: https://github.com/lexfridman/deeptraffic (fetched 2026-08-28T04:05:37.257083+00:00, sha 936c1bb6ebe5)
- Data as of 2026-08-30T08:39:29.467469+00:00.