Jason Lee
@jasondeanlee.bsky.social
Associate Professor at Princeton
Machine Learning Researcher
Machine Learning Researcher
Our new work on scaling laws that includes compute, model size, and number of samples. The analysis involves an extremely fine-grained analysis of online sgd built up over the last 8 years of understanding sgd on simple toy models (tensors, single index models, multi index model)
Excited to announce a new paper with Yunwei Ren, Denny Wu,
@jasondeanlee.bsky.social!
We prove a neural scaling law in the SGD learning of extensive width two-layer neural networks.
arxiv.org/abs/2504.19983
🧵below (1/10)
@jasondeanlee.bsky.social!
We prove a neural scaling law in the SGD learning of extensive width two-layer neural networks.
arxiv.org/abs/2504.19983
🧵below (1/10)
May 5, 2025 at 5:08 PM
Our new work on scaling laws that includes compute, model size, and number of samples. The analysis involves an extremely fine-grained analysis of online sgd built up over the last 8 years of understanding sgd on simple toy models (tensors, single index models, multi index model)
Reposted by Jason Lee
Welcome to the Bluesky account for Stand Up for Science 2025!
Keep an eye on this space for updates, event information, and ways to get involved. We can't wait to see everyone #standupforscience2025 on March 7th, both in DC and locations nationwide!
#scienceforall #sciencenotsilence
Keep an eye on this space for updates, event information, and ways to get involved. We can't wait to see everyone #standupforscience2025 on March 7th, both in DC and locations nationwide!
#scienceforall #sciencenotsilence
February 12, 2025 at 5:04 PM
Welcome to the Bluesky account for Stand Up for Science 2025!
Keep an eye on this space for updates, event information, and ways to get involved. We can't wait to see everyone #standupforscience2025 on March 7th, both in DC and locations nationwide!
#scienceforall #sciencenotsilence
Keep an eye on this space for updates, event information, and ways to get involved. We can't wait to see everyone #standupforscience2025 on March 7th, both in DC and locations nationwide!
#scienceforall #sciencenotsilence
Duck in Vancouver! Mott32
December 11, 2024 at 3:24 AM
Duck in Vancouver! Mott32
Reposted by Jason Lee
“On a log-log plot, my grandmother fits on a straight line.”
-Physicist Fritz Houtermans
There's a lot of truth to this. log-log plots are often abused and can be very misleading
1/5
-Physicist Fritz Houtermans
There's a lot of truth to this. log-log plots are often abused and can be very misleading
1/5
December 3, 2024 at 4:41 AM
“On a log-log plot, my grandmother fits on a straight line.”
-Physicist Fritz Houtermans
There's a lot of truth to this. log-log plots are often abused and can be very misleading
1/5
-Physicist Fritz Houtermans
There's a lot of truth to this. log-log plots are often abused and can be very misleading
1/5
Reposted by Jason Lee
Just put together a starter pack for Deep Learning Theory. Let me know if you'd like to be included or suggest someone to add to the list!
go.bsky.app/2qnppia
go.bsky.app/2qnppia
November 22, 2024 at 9:35 PM
Just put together a starter pack for Deep Learning Theory. Let me know if you'd like to be included or suggest someone to add to the list!
go.bsky.app/2qnppia
go.bsky.app/2qnppia
Zihan Zhang (tinyurl.com/4nks7f9b) is a postdoc with Yuxin Chen, Simon Du, and me.
November 27, 2024 at 8:54 PM
Zihan Zhang (tinyurl.com/4nks7f9b) is a postdoc with Yuxin Chen, Simon Du, and me.
Send your colt open problems to Zihan, with high probability he will solve it!
arxiv.org/abs/2411.17668 Our postdoc zihan slays another COLT open problem! proceedings.mlr.press/v247/kornows...
Anytime Acceleration of Gradient Descent
This work investigates stepsize-based acceleration of gradient descent with {\em anytime} convergence guarantees. For smooth (non-strongly) convex optimization, we propose a stepsize schedule that all...
arxiv.org
November 27, 2024 at 2:33 PM
Send your colt open problems to Zihan, with high probability he will solve it!
arxiv.org/abs/2411.17668 Our postdoc zihan slays another COLT open problem! proceedings.mlr.press/v247/kornows...
Anytime Acceleration of Gradient Descent
This work investigates stepsize-based acceleration of gradient descent with {\em anytime} convergence guarantees. For smooth (non-strongly) convex optimization, we propose a stepsize schedule that all...
arxiv.org
November 27, 2024 at 1:03 PM
arxiv.org/abs/2411.17668 Our postdoc zihan slays another COLT open problem! proceedings.mlr.press/v247/kornows...
What's the point of @perplexity_ai given chatgpt also does search?
November 25, 2024 at 1:06 AM
What's the point of @perplexity_ai given chatgpt also does search?
Yo add me to your starter packs!
November 24, 2024 at 4:23 PM
Yo add me to your starter packs!
Reposted by Jason Lee
Assume that the nodes of a social network can choose between two alternative technologies: B and X.
A node using B receives a benefit with respect to X, but there is a benefit to using the same tech as the majority of your neighbors.
Assume everyone uses X at time t=0. Will they switch to B?
A node using B receives a benefit with respect to X, but there is a benefit to using the same tech as the majority of your neighbors.
Assume everyone uses X at time t=0. Will they switch to B?
November 23, 2024 at 10:48 PM
Assume that the nodes of a social network can choose between two alternative technologies: B and X.
A node using B receives a benefit with respect to X, but there is a benefit to using the same tech as the majority of your neighbors.
Assume everyone uses X at time t=0. Will they switch to B?
A node using B receives a benefit with respect to X, but there is a benefit to using the same tech as the majority of your neighbors.
Assume everyone uses X at time t=0. Will they switch to B?
Reposted by Jason Lee
Starter packs are helpful as well as the twitter import tool chromewebstore.google.com/detail/sky-f...
Sky Follower Bridge - Chrome Web Store
Instantly find and follow the same users from your Twitter follows on Bluesky.
chromewebstore.google.com
November 23, 2024 at 8:36 PM
Starter packs are helpful as well as the twitter import tool chromewebstore.google.com/detail/sky-f...
How do I bulk follow people?
November 23, 2024 at 7:10 PM
How do I bulk follow people?