DL free course
The assistant professor :
Chris G. Willcocks
Department of Computer Science
Durham University
Deep Learning (playlist)The first five lectures are more theoretical, the second half is more applied.
- Lecture 1: Introduction. (slides, video)
- Lecture 2: Mathematical principles and backpropagation. (slides, colab, video)
- Lecture 3: PyTorch programming: coding session. (colab1, colab2, video) - minor issues with audio, but it fixes itself later.
- Lecture 4: Designing models to generalise. (slides, video)
- Lecture 5: Generative models. (slides, desmos, colab, video)
- Lecture 6: Adversarial models. (slides, colab1, colab2, colab3, colab4, video)
- Lecture 7: Energy-based models. (slides, colab, video)
- Lecture 8: Sequential models: by u/samb-t. (slides, colab1, colab2, video)
- Lecture 9: Flow models and implicit networks. (slides, SIREN, GON, video)
- Lecture 10: Meta and manifold learning. (slides, interview, video)
Reinforcement Learning (playlist)This is based on David Silver's course but targeting younger students within a shorter 50min format (missing the advanced derivations) + more examples and Colab code.
- Lecture 1: Foundations. (slides, video)
- Lecture 2: Markov decision processes. (slides, colab, video)
- Lecture 3: OpenAI gym. (video)
- Lecture 4: Dynamic programming. (slides, colab, video)
- Lecture 5: Monte Carlo methods. (slides, colab, video)
- Lecture 6: Temporal-difference methods. (slides, colab, video)
- Lecture 7: Function approximation. (slides, code, video)
- Lecture 8: Policy gradient methods. (slides, code, theory, video)
- Lecture 9: Model-based methods. (slides, video)
- Lecture 10: Extended methods. (slides, atari, video)