lexfridman/mit-deep-learning

Covers hands-on implementations of FFNNs, CNNs, semantic segmentation (DeepLab), and generative models (BigGAN) through interactive Jupyter notebooks executable in Google Colab. Includes the DeepTraffic deep reinforcement learning competition where participants train neural networks to optimize autonomous vehicle behavior in simulated highway environments. Leverages TensorFlow and state-of-the-art pre-trained models to bridge theory with practical applications in computer vision and autonomous systems.

10,422 stars. No commits in the last 6 months.