James Spencer

Excited to share this - can neural networks provide a compact and expressive wavefunction ansatz for quantum chemistry? With wonderful collaborators @pfau, @alexggmatthews at DeepMind and Matthew Foulkes at Imperial. #compchem

I’m beyond thrilled to share that our work on using deep learning to compute excited states of molecules is out today in @ScienceMagazine! This is the first time that deep learning has accurately solved some of the hardest problems in quantum physics. science.org/doi/abs/10.112…

Modifications to the fermionic neural network allow it to tackle studies of a unitary Fermi gas with unrivaled accuracy, suggesting the architecture can also be used to study other strongly correlated systems. go.aps.org/4dNG8Kx @lou_tonny @sutterud @pfau @exp_n

New research shows that learned force fields are ready for use in catalyst discovery. Here, ‘Easy Potentials’ outperform classical quantum chemistry tools on the #OpenCatalyst challenge and find lower energy structures outside of the training set: dpmd.ai/dm-easy-potent… 1/2

✨ Applications for internships are now open✨ Interning at DeepMind offers a unique opportunity to closely collaborate with our team & work on various projects that aim to advance science and benefit humanity. Interested? Learn more: dpmd.ai/internships

Impressive work representing symmetries and antisymmetries in deep networks to model quantum chemical energy calculations.

DeepMind Internship applications open! Note they are accepted on a rolling basis, so apply now! deepmind.com/careers/intern…

Announcing 4 days of "Machine learning for Quantum Simulation" on June 22-23 & June 29-30 @FlatironCCQ - topics include classical ML for quantum sim./VQEs/ML for DFT/quantum tomography etc. Great lineup of speakers, fully virtual, register here to attend! mlqs.org

Isolation, Day 7 DfE leaks abandoned KS2 SAT papers.

Today we are releasing Haiku and RLax, our JAX libraries for neural networks and reinforcement learning. Check them out at github.com/deepmind/haiku and github.com/deepmind/rlax ! #jax #AI #RL #deeplearning #machinelearning

RT to vote for Jess Piasecki! 🇬🇧 Piasecki won the Florence Marathon on her debut at the distance in 2:25:29. Voting closes at 12:00pm CET on Monday 16 December.

Voting STARTS at 7 am. Voting CLOSES at 10 pm. Decide RIGHT NOW when and how you will vote. TELL your plan to a FRIEND. MAKE TIME and VOTE. Even if it's only the LEAST BAD OPTION. VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE VOTE TOMORROW

REGISTER TO VOTE DEADLINE TODAY REGISTER TO VOTE DEADLINE TODAY REGISTER TO VOTE DEADLINE TODAY REGISTER TO VOTE DEADLINE IS TODAY REGISTER TO VOTE DEADLINE TODAY REGISTER TO VOTE DEADLINE TODAY REGISTER TO VOTE DEADLINE 23:59 TODAY gov.uk/register-to-vo…

When I moved to machine learning 11 years ago I thought I was done with quantum physics. Funny how things work out! Work with the exceptional @pfau @exp_n and Matthew Foulkes.

@pfau @alexggmatthews Now I know *almost* nothing about machine learning. A fair exchange?

I should say, I knew nothing about quantum chemistry before I started this project, and I'm incredibly grateful to my collaborators @exp_n, @alexggmatthews at DeepMind and Matthew Foulkes at Imperial. Now I know *almost* nothing about quantum chemistry.

Thrilled to be able to share what I've been working on for the last year - solving the fundamental equations of quantum mechanics with deep learning! arxiv.org/abs/1909.02487