FOLDS seminar: An Information Geometric Understanding of Deep Learning
Zoom link: https://upenn.zoom.us/j/98220304722 I will argue that properties of natural data are what predominantly make deep networks so effective. To that end, I will show that deep networks work well […]
FOLDS seminar: A New Paradigm for Learning with Distribution Shift
Zoom link: https://upenn.zoom.us/j/98220304722 We revisit the fundamental problem of learning with distribution shift, where a learner is given labeled samples from training distribution D, unlabeled samples from test distribution D′ and […]
FOLDS seminar: Theory and practice of LLM quantization
Zoom link: https://upenn.zoom.us/j/98220304722 Modern LLMs process information by repeatedly applying a basic primitive of matrix multiplication. Estimates show that about 60-84% of the energy consumed by LLMs goes into […]
FOLDS seminar: Propagation-of-Chaos in Shallow Neural Networks beyond Logarithmic Time.
Zoom link: https://upenn.zoom.us/j/98220304722 The analysis of gradient-based learning of Neural Networks remains an outstanding challenge, even for the simplest shallow architectures. A powerful mathematical framework that has emerged over recent […]
FOLDS seminar: Heaviside Composite Optimization: A new paradigm of optimization
Zoom link: https://upenn.zoom.us/j/98220304722 A Heaviside function is an indicator function of a semi-infinite interval. A Heaviside composite function is a Heaviside function composed with a multivariate function that may be […]
FOLDS seminar: Algorithmic stability for regression and classification
In a supervised learning setting, a model fitting algorithm is unstable if small perturbations to the input (the training data) can often lead to large perturbations in the output (say, […]
FOLDS Seminar: Positive random walks and positive-semidefinite random matrices
On the real line, a random walk that can only move in the positive direction is very unlikely to remain close to its starting point. After a fixed number of […]
ASSET Seminar: “Reality Checks”
Despite its success, leaderboard chasing has become something researchers dread and mock. When implemented properly and executed faithfully, leaderboard chasing can lead to both faster and easily reproducible progress in […]
ASSET Seminar: “The coverage principle in language models: From pre-training to test-time scaling”
Test-time compute has emerged as a new axis for scaling language model capabilities, yet we lack a principled understanding of this paradigm. What are the right algorithms and trade-offs for […]
ASSET Seminar: “Robust and Uncertainty-Aware Decision Making under Distribution Shifts”
Abstract TBD Zoom: https://upenn.zoom.us/j/92346171614