Talks



Paper session 1 (11.30 - 12.30, 19 Oct 2023 UTC)

Proposed talks

  • In-context interference in Chat-based Large Language Models
    Eric Nuertey Coleman
  • Addressing the challenges of learning in dynamic environments
    Federico Giannini
  • Alleviating Catastrophic Forgetting through Direct Feedback Alignment in Neural Networks
    Sara Folchini

Pre-registration papers

  • Adaptive Hyperparameter Optimization for Continual Learning Scenarios
    Rudy Semola
  • AdaCL: Adaptive Continual Learning
    Elif Ceren Gok Yildirim

Paper session 2 (15.00 - 16.00, 19 Oct 2023 UTC)

Proposed talks

  • Computationally Efficient Continual Learning for Real-World Applications
    Christopher Kanan
  • Lifelong Learning for Evolving Graphs
    Lukas Galke
  • Continual Learning from Demonstration
    Sayantan Auddy, Jakob Hollenstein, Matteo Saveriano, Antonio Rodriguez-sanchez, Justus Piater

Pre-registration papers

  • Implicit Neural Representation as vectorizer for classification task applied to diverse data structures
    Thibault Malherbe

Paper session 3 (18.00 - 19.00, 19 Oct 2023 UTC)

Proposed talks

  • Graphical Neural Activity Threads as an abstraction for spiking neural computation
    Bradley Theilman
  • Dendrites as a biologically inspired approach to overcome catastrophic forgetting
    Jeremy Forest

Pre-registration papers

  • CD-IMM: The Benefits of Domain-based Mixture Models in Bayesian Continual Learning
    Antonio Carta and Daniele Castellana

Paper session 4 (23.30 - 00.30, 19/20 Oct 2023 UTC)

Proposed talks

  • An Analysis of Forgetting in Regularization-Based Continual Learning
    Haoran Li
  • Memory Management Strategies in Replay-Based Continual Learning beyond Reservoir Sampling
    Andrii Krutsylo
  • Developing Strategies for Continual Object Detection with the MMDetection Toolbox
    Angelo Garangau Menezes

Paper session 5 (02.15 - 03.15, 20 Oct 2023 UTC)

Proposed talks

  • Strategies for using pre-trained models as a practical solution for continual learning
    Mark McDonnell

Paper session 6 (07.45 - 08.45, 20 Oct 2023 UTC)

  • Analyzing Continual Learning from a Perspective of Sequential Projections
    Itay Evron, Gon Buzaglo, Edward Moroshko, Nathan Srebro, Daniel Soudry

Pre-registration papers

  • Examining Changes in Internal Representations of Continual Learning Models through Tensor Decomposition
    Nishant Suresh Aswani
  • Two Complementary Perspectives to Continual Learning: Ask Not Only What to Optimize, But Also How
    Timm Hess