Who's Who of AI

Nikhil Garg

290 trust researcher @nkgarg.bsky.social · 2,864 followers
AI research
Why they matter

Researcher with public evidence across AI research.

AI signals
2
past 30d
Sources
1
distinct domains
Discussions
5
past 30d
Latest signal
2d ago
View every signal from Nikhil Garg →
I study algorithms/learning/data applied to democracy/markets/society. Asst. professor at Cornell Tech. https://gargnikhil.com/. Helping building personalized Bluesky research feed: https://bsky.app/profile/paper-feed.bsky.social/feed/preprintdigest
What they're sharing

Articles & links

Very excited to have this out -- @kennypeng.bsky.social 's brainchild in thinking about the implications of sparse representations. He's on the job market, and watch out for some exciting tools we're releasing soon! arxiv.org/abs/2610.05794

A Testable Theory of Atomic Features arxiv.org
AI Weekly's analysis →
  • The paper's 'recovery principle' predicts sparse autoencoders of increasing size recover an increasing prefix of the most prevalent atoms in the training data.
  • Three predictions were tested on SAEs ranging in size from 512 to 131,072, trained on two large embedding models.
  • The authors argue the results contradict conventional wisdom that SAE features are unstable and 'split' as size increases.
Read full analysis →
View on Bluesky · ♥ 11 ↻ 1 ↩ 0 · 2 from the directory shared this · 16d ago
↻ Nikhil Garg reposted
Aaron Roth @aaroth.bsky.social

A world without open problems Here are some that fell today: K-server: arxiv.org/abs/2609.15979 Matroid Secretary: arxiv.org/abs/2609.145... Matrix Spencer: arxiv.org/abs/2609.15025 (Well Matrix Spencer was maybe also a few weeks ago, but who's counting? arxiv.org/abs/2608.288…

The $k$-server conjecture is true arxiv.org
AI Weekly's analysis →
  • A new preprint claims a proof that the work function algorithm achieves competitive ratio k on every metric space, matching the known lower bound.
  • The conjecture was introduced by Manasse, McGeoch and Sleator in 1988; the paper calls it the 'holy grail' of competitive analysis.
  • The prior best general bound for the work function algorithm was 2k-1, due to Koutsoupias and Papadimitriou.
Read full analysis →
 View on Bluesky →
↻ Nikhil Garg reposted
Aaron Roth @aaroth.bsky.social

A world without open problems Here are some that fell today: K-server: arxiv.org/abs/2609.15979 Matroid Secretary: arxiv.org/abs/2609.145... Matrix Spencer: arxiv.org/abs/2609.15025 (Well Matrix Spencer was maybe also a few weeks ago, but who's counting? arxiv.org/abs/2608.288…

The Matroid Secretary Conjecture is True arxiv.org
AI Weekly's analysis →
  • Sahil Singla has posted an arXiv preprint titled 'The Matroid Secretary Conjecture is True,' claiming a full resolution.
  • The algorithm is stated to accept each element of the offline optimum with probability at least 1/4.
  • It requires only the element count in advance and independence-oracle access to arrived elements, not the matroid itself.
Read full analysis →
 View on Bluesky →
↻ Nikhil Garg reposted
Aaron Roth @aaroth.bsky.social

A world without open problems Here are some that fell today: K-server: arxiv.org/abs/2609.15979 Matroid Secretary: arxiv.org/abs/2609.145... Matrix Spencer: arxiv.org/abs/2609.15025 (Well Matrix Spencer was maybe also a few weeks ago, but who's counting? arxiv.org/abs/2608.288…

Matrix Spencer: Eight Standard Deviations Suffice and an Almost-Linear Time Algorithm for Dense Input arxiv.org
AI Weekly's analysis →
  • Zhao Song and Lichen Zhang prove the Matrix Spencer conjecture, showing a signing of spectral discrepancy below 8√n always exists.
  • Their randomized algorithm finds a signing below 12√n using n^{3+o(1)} polylog(1/p) arithmetic operations, matching the size of the dense input up to subpolynomial factors.
  • Previously the conjecture held only under rank, block-diagonal, or Frobenius-norm restrictions; plain random signs delivered only O(√(n log n)).
Read full analysis →
 View on Bluesky →

Paper here: arxiv.org/pdf/2107.07083. Old-ish paper, but newly accepted to Operations Research! With Wes Gurnee, David Shmoys, and David Rothschild.

arxiv.org
View on Bluesky · ♥ 9 ↻ 0 ↩ 2 · 142d ago
↻ Nikhil Garg reposted
Martin Saveski @msaveski.bsky.social

Looking forward to giving a tutorial on 𝗦𝗼𝗰𝗶𝗮𝗹 𝗠𝗲𝗱𝗶𝗮 𝗙𝗲𝗲𝗱 𝗥𝗮𝗻𝗸𝗶𝗻𝗴 𝗔𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀 with @tiziano.bsky.social at @ic2s2.bsky.social! 🕜 𝗧𝗶𝗺𝗲: Tuesday, 1-4 pm 📍 𝗟𝗼𝗰𝗮𝘁𝗶𝗼𝗻: Mansfield (210). 🌐 𝗪𝗲𝗯𝘀𝗶𝘁𝗲: social-media-ais.github.io/ic2s2-26-tut...

Social Media Feed Ranking Algorithms: Guide to Field Experiments social-media-ais.github.io View on Bluesky →
Their network

In Nikhil Garg's orbit

Center = Nikhil Garg. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.

Are you Nikhil Garg? Show it.

Add the Who’s Who of AI badge to your site or bio. It links back to this profile.

Listed in AI Weekly's Who's Who of AI

Markdown: [![Listed in AI Weekly's Who's Who of AI](https://aiweekly.co/modules/custom/aiweekly_whoswho/images/whoswho-badge.svg)](https://aiweekly.co/whos-who/person/nkgarg-bsky-social)