Profile avatar
austinjtripp.bsky.social
(ML ∪ Bayesian optimization ∪ active learning) ∩ (drug discovery) Researcher @valenceai.bsky.social Details: austintripp.ca
24 posts 1,299 followers 226 following
Regular Contributor
Conversation Starter
Posts 15 Comments 15

Can anybody explain to me why so many ML papers study "offline model-based optimization"? This is essentially "1-shot optimization". My main concern is "are there 1-shot optimization problems in real life"? Papers mention "drug discovery (DD)" as an example, but 1-shot DD never happens, no? 😂

submitted 12 days ago • 1 comment

People who are masking are smart for two reasons: 1. They do not want to get brain damage 2. They are not getting brain damage

submitted 19 days ago • 3 comments

My bid screen for ICML position papers is basically: - "Position: ML conference peer review is sh*t" - "Position: Let's abolish conference reviewing" - "Position: C'mon ML reviewers, surely we can do better than *this*" Am I in a "review hate" echo chamber or is everybody else seeing this too? 😶

submitted 26 days ago • 4 comments

Easy to get started on the antiviral challenge! I plan to submit some GP baselines from my PhD work (possibly with a collaborator).

submitted 31 days ago • 0 comments

🏁 The antiviral challenge is live! 🏁 Ready to test your skills on new data? Hosted in partnership with @asapdiscovery.bsky.social and @omsf.io, we've prepared detailed notebooks showcasing how to format your data and submit your solutions. 🧑‍💻

submitted 43 days ago • 1 comment

My PhD thesis is finally online. Thanks @cambridgemlg.bsky.social for a wonderful 4.5 years learning about probabilistic ML 😍 Code: github.com/austint/phd-... Thesis DOI: doi.org/10.17863/CAM...

submitted 44 days ago • 0 comments

A common issue I see in ML, both from ML "experts" and "users", is overly optimistic assumptions. "experts" (people designing algs) usually assume the data is very simple "users" (people using algs) usually assume that algorithms are more robust than they really are Conclusion: always be careful!

submitted 47 days ago • 1 comment

📊 Imagining the Future of ML Evaluation in Drug Discovery Our recent paper discussed the limitations of static leaderboards—they never tell the full story. What if we had a better and easier way of evaluating methods? A vision for the future, in the latest blog 🧵 polarishub.io/blog/imagini...

submitted 70 days ago • 1 comment

Valence is a great place to work- come find me at NeurIPS today if you want to learn more!

submitted 73 days ago • 0 comments

This looks like a really cool competition for small molecule property prediction in both 3D and 2D- great opportunity to work with real data 🚀

submitted 82 days ago • 1 comment

I'm attending NeurIPS next week- reach out if you want to meet for ☕️ I'm particularly interested in meeting: 1. PhD students interested in internships at Valence 2. people working on Bayesian optimization / active learning 3. anyone in tech-bio 4. early-career researchers Details in 🧵 below

submitted 85 days ago • 1 comment

Valence Labs will be co-hosting a TechBio social with @recursionpharma.bsky.social and NVIDIA at #NeurIPS in Vancouver. Join us on Thurs, Dec 12th. RSVP here: lu.ma/biikt7ox Our team will also be at NeurIPS throughout the week. See below for a summary of our papers👇

submitted 91 days ago • 1 comment

The long and ugly story of Cassava Sciences’ supposed Alzheimer’s drug simufilam is finally over. It should have ended well before this.

submitted 92 days ago • 2 comments

Great quick blog post to understand benchmarks in drug discovery. My take: static datasets are probably not enough to prototype algorithms for discovery ‼️

submitted 96 days ago • 0 comments

Now I can share external links without the posts being down regulated - I thought it would share this again!:) I have compiled a list of now 100+ companies in the 'TechBio' space into a fully open database for the community. Find here and please share if you like open.substack.com/pub/harrisbi...

submitted 97 days ago • 2 comments