Section outline

  • The module covers some recent and interesting development and research topics in the field of machine learning. Topics choice is likely to vary at each edition. Example topics include: deep learning for graphs, continual learning, distributed learning, learning-reasoning integration, 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 
     34 14/05/2025
    (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  Software
    - PyDGN: our in-house DLG library
    - PyTorch geometric
    - Deep graph library

    Additional readings
    [57-58] Seminal works on neural networks for graphs
    [59] Recent tutorial paper
    35 15/05/2025
    (14-16)
    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 
    [60] A work on generalizing pooling to graphs
    [61] Probabilistic learning on graphs
    [62] Non-dissipative message passing via neural graph ODEs
    [63] Survey on deep learning for dynamic graphs
    [64] Neural algorithmic reasoning following duality structure in optimization problems

    [65] Seminal work on graph transformers

      20/05/2025 NO LECTURE (due to Giro d'Italia closures)    
    36 21/05/2025
    (16-18)
     (Deep) Reinforcement Learning fundamentals [SD] Sections 19.1-19.3.1, 19.4, 19.5 (no derivation of policy gradient) Additional readings 
    [66] Original Q-Learning algorithm
    [67] Original DQN paper
    [68] Learning with the actor-critic architecture
    [69] A masterpiece paper deriving trust-region policy optimization (technical by worth the read)
    37 22/05/2025
    (14-16)
     Final lecture