7 papers
The Power of Second Order Methods for Sequence Preconditioning
Annie Marsden, Elad Hazan
Sequence prediction methods for linear dynamical systems with long memory, i.e. marginally stable systems, typically achieve regret that grows linearly with the hidden dimension of…
RubiConv -- Efficient Boundary-Respecting Convolutions
Linda Friso, Annie Marsden, Xinyi Chen +4
Convolutional architectures have emerged as powerful alternatives to Transformers for sequence modeling. The primary advantage is that they offer improved theoretical sequence leng…
Spectral Filtering for Complex Linear Dynamical Systems
Elad Hazan, Annie Marsden
We study the problem of learning complex-valued linear dynamical systems (CLDS) with sector-bounded spectrum. This class captures oscillatory and long-memory dynamics arising in si…
Universal Sequence Preconditioning
Annie Marsden, Elad Hazan
We study the problem of preconditioning in sequential prediction. From the theoretical lens of linear dynamical systems, we show that convolving the target sequence corresponds to…
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431
In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…
Isotropic Noise in Stochastic and Quantum Convex Optimization
Annie Marsden, Liam O'Carroll, Aaron Sidford +1
We consider the problem of minimizing a -dimensional Lipschitz convex function using a stochastic gradient oracle. We introduce and motivate a setting where the noise of the sto…