Udaya Ghai
@udayaghai
Followers
133
Following
2K
Media
1
Statuses
31
Applied Scientist @Amazon, PhD in Machine Learning @PrincetonCS
NYC
Joined May 2022
RT @qw3rtman: Discussing "Mind the Gap" tonight at @haizelabs's NYC AI Reading Group with @leonardtang_ and @willccbb. Authors study self-i….
0
9
0
RT @qw3rtman: Still noodling on this, but the generation-verification gap proposed by @yus167 @_hanlin_zhang_ @ShamKakade6 @udayaghai et al….
0
2
0
RT @KempnerInst: 4/26 at 10am:. 'Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models'. @yus167 · @_hanlin_zh….
0
1
0
RT @Abhishek_034: 🎉Excited to present 2 papers at #ICLR2025 in Singapore!. 🧠 Progressive distillation induces an implicit curriculum .📢 Or….
0
10
0
RT @KempnerInst: @_hanlin_zhang_ @johnjvastola @marinkazitnik @nsaphra @ZechenZhang5 4/25 at 10am:. 'How Does Critical Batch Size Scale in….
0
1
0
RT @canondetortugas: Akshay Krishnamurthy and Audrey Huang (@auddery) have written a nice blog post on the intersection of reinforcement le….
0
29
0
RT @HazanPrinceton: Very happy to share the almost-final version of our new online control book:.
0
50
0
RT @NSaunshi: *New ICLR paper* – We introduce a paradigm of *looped models for reasoning*. Main claims.- Reasoning requires depth (via loop….
0
83
0
RT @akhil_bagaria: My 1st last-author paper (joint) will be presented as an Oral at @RealAAAI ! A lot of Option Discovery work (including m….
0
5
0
RT @maxbrudolph: We ran thousands of sweeps to compare RL algos for imperfect information games and found preliminary evidence for the Poli….
0
4
0
RT @realrajpabari: 1/ I'm excited to share a project I worked on at Amazon where we introduce "A shared-revenue Bertrand game" -- a variant….
0
4
0
RT @HazanPrinceton: V excited about a breakthrough in learning linear dynamical systems and sequence prediction, w. my brilliant postdoc,….
0
13
0
RT @ShamKakade6: 1/n In new work, we draw connections between accelerated SGD and various recent optimizers including AdEMAMix, Schedule-Fr….
0
14
0
RT @EugeneVinitsky: We've built a simulated driving agent that we trained on 1.6 billion km of driving with no human data. It is SOTA on e….
0
95
0
RT @tengyuma: RL + CoT works great for DeepSeek-R1 & o1, but: . 1️⃣ Linear-in-log scaling in train & test-time compute.2️⃣ Likely bounded b….
0
108
0
RT @its_dibya: With R1, a lot of people have been asking “how come we didn't discover this 2 years ago?”. Well. 2 years ago, I spent 6 mo….
0
139
0
RT @SPChinchali: Excited to share our new paper in INTERSPEECH '24 on embedding signatures in audio to detect and prevent #deepfake audio.….
0
1
0
RT @XinyiChen2: Together with @HazanPrinceton, Cong, @Zanette_ai, and Nati, we are organizing a long program on reinforcement learning and….
0
6
0