Math and Machine Learning Program Discussion
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesMath and Machine Learning Program Discussion
Loading view.
High-dimensional learning of narrow neural networks
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesMember Seminar Speaker: Hugo Cui, CMSA Title: High-dimensional learning of narrow neural networks Abstract: This talk explores the interplay between neural network architectures and data structure through the lens of high-dimensional asymptotics. We focus on a class of narrow neural networks, namely networks possessing a finite number of hidden units, while operating in high dimensions. In the […]
Math and Machine Learning Program Discussion
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesMath and Machine Learning Program Discussion
Local complexity measures in modern parameterized function classes for supervised learning
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesColloquium Speaker:Â Elisenda Grigsby, Boston College Title: Local complexity measures in modern parameterized function classes for supervised learning Abstract: The parameter space for any fixed architecture of neural networks serves as a proxy during training for the associated class of functions - but how faithful is this representation? For any fixed feedforward ReLU network architecture, it […]
Continuation of solutions of Einstein’s equations
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesGeneral Relativity Seminar Speaker: Oswaldo Vazquez, Northeastern University Title: Continuation of solutions of Einstein's equations Abstract: Klainerman-Rodnianski improved the continuation criterion for the solutions of Einstein's equations proved by Michael Anderson using Kirchoff-Sobolev type parametrix and geometric Littlewood-Paley theory. Using their technique but a new parametrix we prove a continuation condition in the context of […]
Topics in Deep Learning Theory
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesTopics in Deep Learning Theory Eli Grigsby
Math and Machine Learning Program Open Discussion/Tea
Common Room, CMSA 20 Garden Street, Cambridge, MA, United StatesOpen Discussion/Tea
Skein traces and curve counting
Science Center Hall E 1 Oxford Street, Cambridge, MA, United StatesGeometry and Quantum Theory Seminar Speaker: Sunghyuk Park, Harvard CMSA Title: Skein traces and curve counting Abstract: Skein modules are vector space-valued invariants of 3-manifolds describing the space of line defects modulo skein relations (determined by a choice of a ribbon category). When the 3-manifold is S x I for some surface S, the skein […]
Math and Machine Learning Program Discussion
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesMath and Machine Learning Program Discussion
CMSA Q&A Seminar: Laura DeMarco
Common Room, CMSA 20 Garden Street, Cambridge, MA, United StatesCMSA Q&A Seminar Speaker: Laura DeMarco, Harvard University Topic: What is Teichmuller geometry and why is it important?
Topics in Deep Learning Theory
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesTopics in Deep Learning Theory Eli Grigsby
Dolbeault Virasoro algebra and M5 branes
CMSA Room G10 CMSA, 20 Garden Street, Cambridge, MA, United StatesQuantum Field Theory and Physical Mathematics Seminar Speaker: Brian Williams, Boston University Title: Dolbeault Virasoro algebra and M5 branes Abstract: The worldvolume theory on a stack of M5 branes in M-theory is superconformal. We propose a conjecture that in the holomorphic twist of the theory on a stack of M5 branes an infinite-dimensional enhancement of […]