Modeling March Madness with Deep Learning
It’s officially March, and I’ve been biting my nails and having palpitations about the Purdue Boilermakers since sometime in mid-January, so I decided to try...
Recent Posts
Mutual Q-learning
Last night around midnight, I presented a bit of work on the topic of mutual reinforcement learning. We showed that if you allow a couple of AI agents to com...
The Bitter Lesson, and What It Means
We want AI agents that can discover like we can, not which contain what we have discovered.
The Key to the Universe is Just Knowing How to Make Functions Bigger
Every few days, I find myself in awe of the fact that so much progress has been made based on a very simple idea: