Study notebook

Press N to jot a timestamped note or B to pin a moment while watching.

Online Hyperparameter Tuning as an Empirical Approach for Reinforcement Learning

Matthew Vandergrift University of Alberta

MSc Student

57:03·2026·graduate level·Research Seminar

An examination of shifting hyperparameter tuning from an offline process to an open, online procedure within reinforcement learning, presented at the Alberta Machine Intelligence Institute.

AI & ML Reinforcement Learning hyperparameter optimization machine learning empirical methods

In this seminar from the Alberta Machine Intelligence Institute and the Department of Computing Science, Matthew Vandergrift from UofA examines how we can move hyperparameter tuning from a hidden offline procedure towards a open and online procedure.

BIO:

Matthew Vandergrift is a MSc student supervised by Martha White at the University of Alberta.

Comments

Add a comment...Sign in

Related lectures

Machine Learning (CS 229): Lecture 1 - Course Introduction

Andrew Ng · Stanford University

Shared subfield · Reinforcement Learning & machine learning

AI for Materials on Demand: Accelerating Discovery

Aron Walsh · Imperial College London

Shared subfield · Reinforcement Learning & machine learning

The Alberta Plan for AI Research: Continual Reinforcement Learning

Richard S. Sutton · University of Alberta

Shared subfield · Reinforcement Learning

CS 285: Lecture 2, Imitation Learning. Part 1

Sergey Levine · UC Berkeley

Shared subfield · Reinforcement Learning & machine learning

HydroGym: A Reinforcement Learning Platform for Fluid Dynamics

Steven L. Brunton · University of Washington

Shared subfield · Reinforcement Learning

Sequential Optimization of Dynamically Augmented Conditional Value-at-Risk

Alberta Machine Intelligence Institute (Amii)

Shared subfield · Reinforcement Learning