Peter L. Bartlett

Latest

• Random feature amplification: Feature learning and generalization in neural networks
• Benign overfitting without linearity: Neural network classifiers trained by gradient descent for noisy linear data
• The interplay between implicit bias and benign overfitting in two-layer linear networks
• On the theory of reinforcement learning with once-per-episode feedback
• When does gradient descent with logistic loss interpolate using deep networks with smoothed ReLU activations?
• When does gradient descent with logistic loss find interpolating two-layer networks?
• Oracle lower bounds for stochastic gradient Markov chain Monte Carlo methods
• Langevin Monte Carlo without smoothness
• OSOM: A simultaneously optimal algorithm for multi-armed and linear contextual bandits
• Is there an analog of Nesterov acceleration for MCMC?
• Convergence rates for Langevin Monte Carlo in the nonconvex setting
• Online learning with kernel losses
• On the theory of variance reduction for stochastic gradient Monte Carlo
• Underdamped Langevin MCMC: A non-asymptotic analysis
• Alternating minimization for dictionary learning: Local convergence guarantees