@anaymehrotra.bsky.social
PhD candidate @ Yale | Undergrad @ IITK | anaymehrotra.com
Learning Theory, Missing Data, Generation
Learning Theory, Missing Data, Generation
Nitpick: Berry–Esseen :P
July 29, 2025 at 6:11 AM
Nitpick: Berry–Esseen :P
Organized with a stellar team of co-organizers – Andrew Ilyas (aifi.bsky.social), Alkis Kalavasis, and Manolis Zampetakis
https://aifi.bsky.social
July 28, 2025 at 5:14 PM
Organized with a stellar team of co-organizers – Andrew Ilyas (aifi.bsky.social), Alkis Kalavasis, and Manolis Zampetakis
A fantastic lineup of speakers/panelists: Ahmad Beirami (abeirami.bsky.social), Surbhi Goel (surbhigoel.bsky.social���), Steve Hanneke, Chris Harshaw, Amin Karbasi (aminkarbasi.bsky.social), Samory Kpotufe, Chara Podimata (charapod.bsky.social), …
https://abeirami.bsky.social
July 28, 2025 at 5:14 PM
A fantastic lineup of speakers/panelists: Ahmad Beirami (abeirami.bsky.social), Surbhi Goel (surbhigoel.bsky.social���), Steve Hanneke, Chris Harshaw, Amin Karbasi (aminkarbasi.bsky.social), Samory Kpotufe, Chara Podimata (charapod.bsky.social), …
More info: reliablemlworkshop.github.io
Reliable ML from Unreliable Data — NeurIPS 2025 Workshop
ReliableMLWorkshop.github.io
July 28, 2025 at 5:14 PM
More info: reliablemlworkshop.github.io
Paper ➜ arxiv.org/abs/2504.07133
Can SGD Select Good Fishermen? Local Convergence under Self-Selection Biases and Beyond
We revisit the problem of estimating $k$ linear regressors with self-selection bias in $d$ dimensions with the maximum selection criterion, as introduced by Cherapanamjeri, Daskalakis, Ilyas, and Zamp...
arxiv.org
May 9, 2025 at 8:11 PM
Paper ➜ arxiv.org/abs/2504.07133
→ CDIZ’23: Runtime ≈ poly(d) · exp(k/ε)
→ Gaitonde–Mossel’24: Runtime ≈ poly(d) / εᵏ (+ also have optimal sample complexity)
→ Ours: Runtime ≈ poly(d) / ε² via warm start + SGD
Key insight: Recast self-selection as regression with coarse (aka rounded) labels
→ Gaitonde–Mossel’24: Runtime ≈ poly(d) / εᵏ (+ also have optimal sample complexity)
→ Ours: Runtime ≈ poly(d) / ε² via warm start + SGD
Key insight: Recast self-selection as regression with coarse (aka rounded) labels
May 9, 2025 at 8:11 PM
→ CDIZ’23: Runtime ≈ poly(d) · exp(k/ε)
→ Gaitonde–Mossel’24: Runtime ≈ poly(d) / εᵏ (+ also have optimal sample complexity)
→ Ours: Runtime ≈ poly(d) / ε² via warm start + SGD
Key insight: Recast self-selection as regression with coarse (aka rounded) labels
→ Gaitonde–Mossel’24: Runtime ≈ poly(d) / εᵏ (+ also have optimal sample complexity)
→ Ours: Runtime ≈ poly(d) / ε² via warm start + SGD
Key insight: Recast self-selection as regression with coarse (aka rounded) labels
Self-selection arises when agents face options r_1, …,r_k and strategically choose the one to, e.g., maximize the reward
The seminal work of Roy (1951) introduced learning with self-selection
Even identification for unknown-index variant is nascent, starting w/ CDIZ [STOC’23]
The seminal work of Roy (1951) introduced learning with self-selection
Even identification for unknown-index variant is nascent, starting w/ CDIZ [STOC’23]
May 9, 2025 at 8:11 PM
Self-selection arises when agents face options r_1, …,r_k and strategically choose the one to, e.g., maximize the reward
The seminal work of Roy (1951) introduced learning with self-selection
Even identification for unknown-index variant is nascent, starting w/ CDIZ [STOC’23]
The seminal work of Roy (1951) introduced learning with self-selection
Even identification for unknown-index variant is nascent, starting w/ CDIZ [STOC’23]
A research paper?
April 29, 2025 at 7:44 PM
A research paper?
Growing list of contributors/researchers
Cornell/MIT: Jon Kleinberg, @sendhil.bsky.social
Duke: Fan Wei
Stanford: Moses Charikar, Chirag Pabbaraju, Charlotte Peale, Omer Reingold
U Michigan: Jiaxun Li, @vkraman.bsky.social , Ambuj Tewari
Yale: Alkis Kalavasis, Anay mehrotra, Grigoris Velegkas
Cornell/MIT: Jon Kleinberg, @sendhil.bsky.social
Duke: Fan Wei
Stanford: Moses Charikar, Chirag Pabbaraju, Charlotte Peale, Omer Reingold
U Michigan: Jiaxun Li, @vkraman.bsky.social , Ambuj Tewari
Yale: Alkis Kalavasis, Anay mehrotra, Grigoris Velegkas
April 21, 2025 at 11:53 PM
Growing list of contributors/researchers
Cornell/MIT: Jon Kleinberg, @sendhil.bsky.social
Duke: Fan Wei
Stanford: Moses Charikar, Chirag Pabbaraju, Charlotte Peale, Omer Reingold
U Michigan: Jiaxun Li, @vkraman.bsky.social , Ambuj Tewari
Yale: Alkis Kalavasis, Anay mehrotra, Grigoris Velegkas
Cornell/MIT: Jon Kleinberg, @sendhil.bsky.social
Duke: Fan Wei
Stanford: Moses Charikar, Chirag Pabbaraju, Charlotte Peale, Omer Reingold
U Michigan: Jiaxun Li, @vkraman.bsky.social , Ambuj Tewari
Yale: Alkis Kalavasis, Anay mehrotra, Grigoris Velegkas
Missed the tag above: sendhil.bsky.social
Sendhil Mullainathan (@sendhil.bsky.social)
sendhil.bsky.social
November 25, 2024 at 7:32 PM
Missed the tag above: sendhil.bsky.social