What We've Learned to Control
I’m giving a keynote address at the virtual IFAC congress this July, and I submitted an abstract that forces me to reflect on the current state of research at the intersection of machine learning and... Continue
The Uncanny Valley of Virtual Conferences
We wrapped up two amazing days of L4DC 2020 last Friday. It’s pretty wild to watch this community grow so quickly: starting as a workshop at CDC 2018, the conference organizers put together an inaugural... Continue
You Cannot Serve Two Masters: The Harms of Dual Affiliation
Facebook would like to have computer science faculty in AI committed to work 80% of their time in industrial jobs and 20% of their time at their university. They call this scheme “co-employment” or “dual... Continue
Towards Actionable Intelligence
I’m going to close my outsider’s tour of Reinforcement Learning by announcing the release of a short survey of RL that coalesces my views from the perspectives of continuous control. Though the RL and controls... Continue
Coarse-ID Control
This is the thirteenth part of “An Outsider’s Tour of Reinforcement Learning.” Part 14 is here. Part 12 is here. Part 1 is here. Can poor models be used in control loops and still achieve... 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
Newer