1 citations · 2 across the 12 of their papers we have counts for
3 papers · 1 filter
Faster Inference of Flow-Based Generative Models via Improved Data-Noise Coupling
Aram Davtyan, Leello Tadesse Dadi, Volkan Cevher +1
Conditional Flow Matching (CFM), a simulation-free method for training continuous normalizing flows, provides an efficient alternative to diffusion models for key tasks like image…
KOALA++: Efficient Kalman-Based Optimization with Gradient-Covariance Products
Zixuan Xia, Aram Davtyan, Paolo Favaro
We propose KOALA++, a scalable Kalman-based optimization algorithm that explicitly models structured gradient uncertainty in neural network training. Unlike second-order methods, w…
KOALA: A Kalman Optimization Algorithm with Loss Adaptivity
Aram Davtyan, Sepehr Sameni, Llukman Cerkezi +3
Optimization is often cast as a deterministic problem, where the solution is found through some iterative procedure such as gradient descent. However, when training neural networks…