Daniel Freeman

I'm a Research Engineer and lead on the Brax team at Google. Prior to Google, I received my PhD in Physics from the University of California at Berkeley, where I studied topological quantum error correction and tensor network methods. While still interested in Quantum Information Science, since joining Google, I've dabbled in metalearning, model based reinforcement learning, rigid body physics for simulation, and large language model-ing. I'm particularly interested in algorithms for learning representations that are generally useful for tasks, as well as algorithms that learn to learn representations that are generally useful, and algorithms that learn to learn to learn, and so on. Nowadays, most of my time is spent developing Brax, a differentiable, massively parallelizable physics engine that enables greatly accelerated reinforcement learning of physically simulated environments. In my free time, I'm interested in the art of story telling--be that through books, film, games, web serials or anything else.