Non-convex learning, system identification, and more! | Prof. Andrew Lamperski | MSR Colloquium
In this talk, Dr. Lamperski will first examine the convergence of Langevin algorithms for machine learning and system identification problems with constraints.
Much of machine learning fits model parameters to data via optimization, typically via some variation of stochastic gradient descent. However, in many cases, such as neural network regression, the loss functions are non-convex and stochastic gradient descent can get stuck in local minima, if it even converges. Langevin methods augment standard gradient-based methods with additive noise. In the case of unconstrained problems, it is well-understood how the additive noise helps the algorithm escape undesirable minima. However, many neural network regression and probabilistic estimation problems require constraints.
Official Event Page: https://cse.umn.edu/mnri/events/robot...
Minnesota Robotics Institute: https://cse.umn.edu/MnRI
MS in Robotics Program: z.umn.edu/MSROverview
Facebook: / minnesotaroboticsinstitute
Twitter: / umn_mnri
Instagram: / umn_mnri