Erin Grant

@eringrant.me

5578 Followers 1521 Following

Assistant Professor / Faculty Fellow @nyudatascience.bsky.social studying cognition in mind & brain with neural nets, Bayes, and other tools (eringrant.github.io). elsewhere: sigmoid.social/@eringrant, twitter.com/ermgrant @ermgrant

Cathy gives the opening remarks.
Function-representation dissociation in ReLU networks. (A-B) MNIST representations before/after prediction-preserving reparametrisation. (C) RSM after function-preserving reparametrisation. (D-E) Performance under input/parameter noise for different solution types.
Hidden-layer representations for a semantic hierarchy task. (A) Task structure. (B) Input/target encoding. (C-E) Hidden representations and representational similarity matrices for task-agnostic (C: LSS) vs. task-specific (D: MRNS, E: MWNS) solutions.
The solution manifold. (A) Solution manifold for a 3-parameter linear network, showing GLS and constrained LSS, MRNS, and MWNS solutions. (B-E) Input/output weight relationships and parametrisation structure for each solution type.
Task solution hierarchy defined by implicit regularisation objectives.
Example of a failure case. (A) A random walk on the solution manifold of a two-layer linear network reveals that weights can change continuously, inducing changes in the (B) network parametrisation and thus the (C) hidden-layer representations, while preserving the (D) network output.
The three questions of the GAC? 1. What should benchmarks measure? 2. What should the goals of a benchmark be? 3. How should benchmarks be structured?
Speakers and organizers of the GAC debate. Time and location of the GAC debate: 5 PM in Room C1.03.
Leon presenting poster at the NeurIPS poster session to an intrigued audience