7 papers
Convergence of Continual Learning in Homogeneous Deep Networks
Matan Schliserman, Gon Buzaglo, Itay Evron +1
We characterize weakly regularized continual classification in homogeneous models as sequential projections onto task margin sets. This result generalizes prior analyses restricted…
Mirror Descent Beyond Euclidean Stability: An Exponential Separation in Initialization Sensitivity
Shira Vansover-Hager, Matan Schliserman, Ofir Schlisselberg +1
Mirror Descent (MD) extends Gradient Descent (GD) beyond Euclidean geometry and has recently reappeared as a lens for KL-regularized policy optimization in reinforcement learning a…
Flat Minima and Generalization: Insights from Stochastic Convex Optimization
Matan Schliserman, Shira Vansover-Hager, Tomer Koren
Understanding the generalization behavior of learning algorithms is a central goal of learning theory. A recently emerging explanation is that learning algorithms are successful in…
From Continual Learning to SGD and Back: Better Rates for Continual Linear Models
Itay Evron, Ran Levinstein, Matan Schliserman +4
We study the common continual learning setup where an overparameterized model is sequentially fitted to a set of jointly realizable tasks. We analyze forgetting, defined as the los…
Optimal Rates in Continual Linear Regression via Increasing Regularization
Ran Levinstein, Amit Attia, Matan Schliserman +4
We study realizable continual linear regression under random task orderings, a common setting for developing continual learning theory. In this setup, the worst-case expected loss…
Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime
Amit Attia, Matan Schliserman, Uri Sherman +1
We study population convergence guarantees of stochastic gradient descent (SGD) for smooth convex objectives in the interpolation regime, where the noise at optimum is zero or near…