Search Results for author: Goutham Rajendran

Found 10 papers, 3 papers with code

On the Origins of Linear Representations in Large Language Models

no code implementations6 Mar 2024 Yibo Jiang, Goutham Rajendran, Pradeep Ravikumar, Bryon Aragam, Victor Veitch

To that end, we introduce a simple latent variable model to abstract and formalize the concept dynamics of the next token prediction.

Language Modelling Large Language Model

Learning Linear Causal Representations from Interventions under General Nonlinear Mixing

no code implementations NeurIPS 2023 Simon Buchholz, Goutham Rajendran, Elan Rosenfeld, Bryon Aragam, Bernhard Schölkopf, Pradeep Ravikumar

We study the problem of learning causal representations from unknown, latent interventions in a general setting, where the latent distribution is Gaussian but the mixing function is completely general.

counterfactual

Nonlinear Random Matrices and Applications to the Sum of Squares Hierarchy

no code implementations9 Feb 2023 Goutham Rajendran

In this work, we analyze the performance of the SoS hierarchy on fundamental problems stemming from statistics, theoretical computer science and statistical physics.

Combinatorial Optimization

Concentration of polynomial random matrices via Efron-Stein inequalities

no code implementations6 Sep 2022 Goutham Rajendran, Madhur Tulsiani

Using our general framework, we derive bounds for "sparse graph matrices", which were obtained only recently by Jones et al. [FOCS 2021] using a nontrivial application of the trace power method, and was a core component in their work.

Tensor Networks

Analyzing Robustness of End-to-End Neural Models for Automatic Speech Recognition

1 code implementation17 Aug 2022 Goutham Rajendran, Wei Zou

Therefore, the models we develop for various tasks should be robust to such kinds of noisy data, which led to the thriving field of robust machine learning.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +1

Identifiability of deep generative models without auxiliary information

no code implementations20 Jun 2022 Bohdan Kivva, Goutham Rajendran, Pradeep Ravikumar, Bryon Aragam

We prove identifiability of a broad class of deep latent variable models that (a) have universal approximation capabilities and (b) are the decoders of variational autoencoders that are commonly used in practice.

Structure learning in polynomial time: Greedy algorithms, Bregman information, and exponential families

no code implementations NeurIPS 2021 Goutham Rajendran, Bohdan Kivva, Ming Gao, Bryon Aragam

Greedy algorithms have long been a workhorse for learning graphical models, and more broadly for learning statistical models with sparse structure.

Learning latent causal graphs via mixture oracles

1 code implementation NeurIPS 2021 Bohdan Kivva, Goutham Rajendran, Pradeep Ravikumar, Bryon Aragam

We study the problem of reconstructing a causal graphical model from data in the presence of latent variables.

Cannot find the paper you are looking for? You can Submit a new open access paper.