no code implementations • 8 Mar 2024 • Xun Tang, Holakou Rahmanian, Michael Shavlovsky, Kiran Koshy Thekumparampil, Tesi Xiao, Lexing Ying
We derive the corresponding entropy regularization formulation and introduce a Sinkhorn-type algorithm for such constrained OT problems supported by theoretical guarantees.
no code implementations • 20 Jan 2024 • Xun Tang, Michael Shavlovsky, Holakou Rahmanian, Elisa Tardini, Kiran Koshy Thekumparampil, Tesi Xiao, Lexing Ying
To achieve possibly super-exponential convergence, we present Sinkhorn-Newton-Sparse (SNS), an extension to the Sinkhorn algorithm, by introducing early stopping for the matrix scaling steps and a second stage featuring a Newton-type subroutine.
no code implementations • 14 Oct 2023 • Liang Zhang, Bingcong Li, Kiran Koshy Thekumparampil, Sewoong Oh, Niao He
The widespread practice of fine-tuning large language models (LLMs) on domain-specific data faces two major challenges in memory and privacy.
no code implementations • 31 Jul 2023 • Charlie Hou, Kiran Koshy Thekumparampil, Michael Shavlovsky, Giulia Fanti, Yesh Dattatreya, Sujay Sanghavi
On tabular data, a significant body of literature has shown that current deep learning (DL) models perform at best similarly to Gradient Boosted Decision Trees (GBDTs), while significantly underperforming them on outlier data.
no code implementations • 1 Jun 2022 • Liang Zhang, Kiran Koshy Thekumparampil, Sewoong Oh, Niao He
We provide a general framework for solving differentially private stochastic minimax optimization (DP-SMO) problems, which enables the practitioners to bring their own base optimization algorithm and use it as a black-box to obtain the near-optimal privacy-loss trade-off.
no code implementations • 19 Jan 2022 • Kiran Koshy Thekumparampil, Niao He, Sewoong Oh
We also provide a direct single-loop algorithm, using the LPD method, that achieves the iteration complexity of $O(\sqrt{\frac{L_x}{\varepsilon}} + \frac{\|A\|}{\sqrt{\mu_y \varepsilon}} + \sqrt{\frac{L_y}{\varepsilon}})$.
no code implementations • 18 May 2021 • Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
We show that, for a constant subspace dimension MLLAM obtains nearly-optimal estimation error, despite requiring only $\Omega(\log d)$ samples per task.
no code implementations • NeurIPS 2020 • Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
Further, instead of a PO if we only have a linear minimization oracle (LMO, a la Frank-Wolfe) to access the constraint set, an extension of our method, MOLES, finds a feasible $\epsilon$-suboptimal solution using $O(\epsilon^{-2})$ LMO calls and FO calls---both match known lower bounds, resolving a question left open since White (1993).
2 code implementations • NeurIPS 2019 • Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
This paper studies first order methods for solving smooth minimax optimization problems $\min_x \max_y g(x, y)$ where $g(\cdot,\cdot)$ is smooth and $g(x,\cdot)$ is concave for each $x$.
1 code implementation • 14 Jun 2019 • Zinan Lin, Kiran Koshy Thekumparampil, Giulia Fanti, Sewoong Oh
Disentangled generative models map a latent code vector to a target space, while enforcing that a subset of the learned latent codes are interpretable and associated with distinct properties of the target distribution.
no code implementations • 9 Jun 2019 • Kiran Koshy Thekumparampil, Sewoong Oh, Ashish Khetan
Matching the performance of conditional Generative Adversarial Networks with little supervision is an important task, especially in venturing into new domains.
2 code implementations • NeurIPS 2018 • Kiran Koshy Thekumparampil, Ashish Khetan, Zinan Lin, Sewoong Oh
When the distribution of the noise is known, we introduce a novel architecture which we call Robust Conditional GAN (RCGAN).