Speaker: Mengye Ren Title: Lifelong and Human-like Learning in Foundation Models Abstract: Real-world agents, including humans, learn from online, lifelong experiences. However, today’s foundation models primarily acquire knowledge through offline, iid learning, while relying on in-context learning for most online adaptation. It is crucial to equip foundation models with lifelong and human-like learning abilities to enable more flexible…

NSF Site Visit - The NSF team will evaluate the progress and achievements of ARNI’s projects to date and provide recommendations to steer future directions and funding for the project. If you are interested in learning more about ARNI over-all, join this Zoom link from 9am to 12pm or 2pm to 4:30pm.

Title: Large-scale pretraining on neural data allows for transfer across individuals, tasks and species Abstract: As neuroscience datasets grow in size and complexity, integrating diverse data sources to achieve a comprehensive understanding of brain function presents both an opportunity and a challenge. In this talk, I will introduce our approach to developing a multi-source foundation model for…

Speaker: Haozhe Shan Title: A theory of continual learning in deep neural networks: task relations, network architecture and learning procedure Abstract: Imagine listening to this talk and afterwards forgetting everything else you’ve ever learned. This absurd scenario would be commonplace if the brain could not perform continual learning (CL) – acquiring new skills and knowledge without…

Title: Bounded optimality: A cognitive perspective on neural computation with resource limitations

Title: Neuronal energy consumption: basic measures and trade-offs, and their effects on efficiency Zoom: https://columbiauniversity.zoom.us/j/98299154214?pwd=1J3J0lEpF6XdqHkHy02c7LuD6xUWx2.1

Title: Taskonomy: Disentangling Task Transfer Learning Abstract: TBD  Link: http://taskonomy.stanford.edu/taskonomy_CVPR2018.pdf

Title: Meta-learning for more powerful behavioral modeling Abstract: Two modeling paradigms have historically been in tension: Bayesian models provide an elegant way to incorporate prior knowledge, but they make simplifying and constraining assumptions; on the other hand, neural networks provide great modeling flexibility, but they make it difficult to incorporate prior knowledge. Here I describe how to…

Title: A Practitioner’s Guide to Continual Multimodal Pretraining Reading: https://arxiv.org/pdf/2408.1447 Zoom: https://columbiauniversity.zoom.us/j/97176853843?pwd=VLZdh6yqHBcOQhdf816lkN5ByIpIsF.1

Title: Target Learning rather than Backpropagation Explains Learning in the Mammalian Neocortex Abstract: Modern computational neuroscience presents two competing hypotheses for hierarchical learning in the neocortex: (1) deep learning-inspired approximations of the backpropagation algorithm, where neurons adjust synapses to minimize error, and (2) target learning algorithms, where neurons reduce the feedback required to achieve a desired activity.…

Title: Meta Continual Learning Revisited: Implicitly Enhancing Online Hessian Approximation via Variance Reduction Reading: https://openreview.net/pdf?id=TpD2aG1h0D Zoom: https://columbiauniversity.zoom.us/j/97176853843?pwd=VLZdh6yqHBcOQhdf816lkN5ByIpIsF.1

General Agenda October 21st, Day 1 from 8:45am to 5pm Breakfast and Lunch Provided Opening 3 Keynote Speakers from ARNI Faculty Research Brainstorming and Discussions Project/Student Poster Session Education and Broader Impact Discussions October 22nd, Day 2 from 9am to 1pm Breakfast and Lunch Provided 1 Keynote Speaker Brainstorming and Discussion on Collaborations & Knowledge…