An Outsider's Tour of Reinforcement Learning

Continue

Coarse-ID Control

Can poor models be used in control loops and still achieve near-optimal performance? In recent posts, we’ve seen the answer is certainly “maybe.” Nominal control could learn a poor model of the double-integrator with 10... Continue

Lost Horizons

This is the twelfth part of “An Outsider’s Tour of Reinforcement Learning.” Part 13 is here. Part 11 is here. Part 1 is here. This series began by describing a view of reinforcement learning as... Continue

Catching Signals That Sound in the Dark

This is the eleventh part of “An Outsider’s Tour of Reinforcement Learning.” Part 12 is here. Part 10 is here. Part 1 is here. The essence of reinforcement learning is using past data to enhance... Continue

The Best Things in Life Are Model Free

This is the tenth part of “An Outsider’s Tour of Reinforcement Learning.” Part 11 is here. Part 9 is here. Part 1 is here. Though I’ve spent the last few posts casting shade at model-free... Continue

The Ethics of Reward Shaping

I read three great articles over the weekend by Renee DiResta, Chris Wiggins, and Janelle Shane that touched on a topic that’s been troubling me: In machine learning, we take our cost functions for granted,... Continue

Benchmarking Machine Learning with Performance Profiles

A common sticking point in contemporary reinforcement learning is how to evaluate performance on benchmarks. For a general purpose method, we’d like to demonstrate aptitude on a wide selection of test problems with minimal special... Continue

Clues for Which I Search and Choose

This is the ninth part of “An Outsider’s Tour of Reinforcement Learning.” Part 10 is here. Part 8 is here. Part 1 is here. Before we leave these model-free chronicles behind, let me turn to... Continue

Updates on Policy Gradients

This is the eighth part of “An Outsider’s Tour of Reinforcement Learning.” Part 9 is here. Part 7 is here. Part 1 is here. I’ve been swamped with a bit of a travel binge and... Continue

A Model, You Know What I Mean?

This is the seventh part of “An Outsider’s Tour of Reinforcement Learning.” Part 8 is here. Part 6 is here. Part 1 is here. The role of models in reinforcement learning remains hotly debated. Model-free... Continue