Title: Brain-like learning with exponentiated gradients Abstract: Computational neuroscience relies on gradient descent (GD) for training artificial neural network (ANN) models of the brain. The advantage of GD is that it is effective at learning difficult tasks. However, it produces ANNs that are a poor phenomenological fit to biology, making them less relevant as models…

Continuation of the prior meeting. Zoom: Upon request @[email protected]

ARNI will host an informal brainstorming session on July 15th (ZI Education Lab) focused on developing AI and neuroscience/cognitive science activities for K–12 students. The goal is to create engaging ways to help young learners better understand the brain and artificial intelligence. Trainees are encouraged to attend—if you're interested in making an impact on youth…

Title: Meta-dynamical state space modeling for integrative neural data analysis Abstract: Uncovering the organizing principles of neural systems requires integrating information across diverse datasets—each alone offering a limited view and signal-to-noise ratio, but together revealing coherent dynamical structures. We present a meta-dynamical state-space modeling framework that learns a shared solution space of neural dynamics from…

Title: From 2D Chips to 3D Brains Abstract:  Artificial intelligence (AI) realizes a synaptocentric conception of the learning brain with dot-products and advances by performing twice as many multiplications every two months. But the semiconductor industry tiles twice as many multipliers on a chip only every two years. Moreover, the returns from tiling these multipliers…

Title and Abstracts:  1st Speaker: Vinam Arora, UPenn Title: Know Thyself by Knowing Others: Learning Neuron Identity from Population Context Abstract: Identifying the functional identity of individual neurons is essential for interpreting circuit dynamics, yet remains a major challenge in large-scale in vivo recordings where anatomical and molecular labels are often unavailable. Here we introduce…

Title: Neural circuit mechanisms of value-based decision-making Abstract:  The value of the environment determines animals’ motivational states and sets expectations for error-based learning. But how are values computed? We developed a novel temporal wagering task with latent structure, and used high-throughput behavioral training to obtain well-powered behavioral datasets from hundreds of rats that learned the…

Zoom link: Upon request @[email protected]

Title: Fast and Flexible Group Decision-Making Abstract: A wide range of animals live and move in groups. Many animals do better in groups than alone when, for example, foraging for food, migrating, and avoiding predators. A key to group success is social interaction. Less well understood is how a group, with no centralized control, is…

Monday (9/15) the Continual Learning Group will have a presentation from group member Yunfan Zhang.  Yunfan will be sharing his ongoing work on developing a continual learning benchmark based on deriving up-to-date facts from news over time. Date: Monday, September 15 Time: 3-4pm Room: CSB 480 Zoom: upon request @ [email protected]

Title and Abstract: TBD Zoom: Meeting ID: 993 3345 6502 Passcode: Upon request @ [email protected]