Section outline
-
The module covers some recent and interesting development and research topics in the field of generative and deep learning. Topics choice is likely to vary at each edition. Example topics include: deep learning for graphs, reinforcement learning, continual learning, edgeAI, lerning beyond backpropagation, neural networks inspired by dynamical systems, The module concludes with a final lecture which discusses the course content retrospectively and details the exam modalities, topics and deadlines.
Date Topic References Additional Material 32 13/05/2026
(16-18)Deep learning for graphs I: Fundamentals
learning with structured data, learning tasks on graphs, message-passing architectures, survey of foundational models for graphs[SD] Chapter 13 Additional readings
[46-47] Seminal works on neural networks for graphs
[48] Recent tutorial paperSoftware
- PyTorch geometric
- Deep graph library33 14/05/2026
(11-13)Deep learning for graphs II: advanced topics
graph convolutional networks, graph pooling, generative learning on graphs, probabilistic graph models, non-dissipative graph message passing, neural algorithmic reasoning[SD] Chapter 13 Additional readings
[49] Probabilistic learning on graphs
[50] Non-dissipative message passing via neural graph ODEs
[51] Survey on deep learning for dynamic graphs
[52] Neural algorithmic reasoning following duality structure in optimization problems[53] Seminal work on graph transformers
34 19/05/2026
(11-13)(Deep) Reinforcement Learning fundamentals
RL problem, Markov decision processes, policy/value iteration, Q-learning, policy learning, elements of deep RL
[SD] Sections 19.1-19.3.1, 19.4, 19.5 (no derivation of policy gradient) Additional readings
[54] Original Q-Learning algorithm
[55] Original DQN paper
[56] Learning with the actor-critic architecture
[57] A masterpiece paper deriving trust-region policy optimization (technical by worth the read)35 20/05/2026
(16-18)Final lecture