🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

I am pleased to announce that the camera ready version of my new textbook, "Probabilistic Machine Learning: An Introduction", is finally available from probml.ai. Hardcopies will be available from MIT Press in Feb 2022.

I wanted to share some recent work on generalisation: "On the Implicit Biases of Architecture & Gradient Descent", with Yisong Yue. Feedback is welcomed :) arxiv.org/abs/2110.04274 (1/6)

Looks like an awesome opportunity!

We look forward to receiving your paper submissions to the 4th Symposium on Advances in Approximate Bayesian Inference (AABI)! Check out our website for updates: approximateinference.org (with Ruqi Zhang, Vincent Fortuin @vincefort.bsky.social, Theofanis Karaletsos, yingzhen)

I've had the joy of collaborating with Stephan over the years, and he's a great advisor with a wonderful group. UC Irvine is also a great place to be, both academically and weather-wise. I once gave a guest lecture, then went to the beach 😎

I'll be presenting Iterative Amortized Policy Optimization during poster session 1 tomorrow (spot E2). It's a new way to perform policy optimization based on ideas from probabilistic inference. Stop by and say hello if you're interested in learning more!

In case you missed it...lots has been written about predictive coding, but rarely with such clarity as in this article by Joseph Marino. It traces the historical roots in cybernetics, and explores connections to current ML. Really amazing! arxiv.org/abs/2011.07464

"Object Representations as Fixed Points: Training Iterative Refinement Algorithms with Implicit Differentiation" is now out on arXiv! w/ my advisors Griffiths Computational Cognitive Science Lab, Sergey Levine arxiv: arxiv.org/abs/2207.00787 web: sites.google.com/view/implicit-… youtube: youtu.be/hxGNnfZRuB8

I'm now on the job market, looking for postdoc and industry positions. Interested in (1) developing AI for automatically modeling and manipulating systems (2) developing AI for powering next generation human-computer interfaces. If you know any opportunities, please let me know!

I’m at NeurIPS for the week. Looking forward to seeing everyone!

Anisotropic diffusion uses space-varying conductivity to slow down diffusion near edges, resulting in non-isotropic diffusion behavior. en.wikipedia.org/wiki/Anisotrop…

I’m at NeurIPS 2023 for the week! Looking forward to seeing familiar faces and meeting new people

Excited to announce this publicly! Continuing the tradition of Google DeepMind’s research using video game environments in this new era of AI