38 citations · 247 across the 108 of their papers we have counts for
14 papers · 1 filter
Optimal Scalar Quantization for Matrix Multiplication: Closed-Form Density and Phase Transition
Calvin Ang, Sungyoon Kim, Mert Pilanci
We study entrywise scalar quantization of two matrices prior to multiplication. Given and , we quantize entries of and independentl…
An Information-Theoretic Perspective on LLM Tokenizers
Mete Erdogan, Abhiram Gorle, Shubham Chandak +2
Large language model (LLM) tokenizers act as structured compressors: by mapping text to discrete token sequences, they determine token count (and thus compute and context usage) an…
Iterative Sketching for Secure Coded Regression
Neophytos Charalambides, Hessam Mahdavifar, Mert Pilanci +1
Linear regression is a fundamental and primitive problem in supervised machine learning, with applications ranging from epidemiology to finance. In this work, we propose methods fo…
Gradient Coding with Iterative Block Leverage Score Sampling
Neophytos Charalambides, Mert Pilanci, Alfred Hero
We generalize the leverage score sampling sketch for -subspace embeddings, to accommodate sampling subsets of the transformed data, so that the sketching approach is approp…
Securely Aggregated Coded Matrix Inversion
Neophytos Charalambides, Mert Pilanci, Alfred Hero
Coded computing is a method for mitigating straggling workers in a centralized computing network, by using erasure-coding techniques. Federated learning is a decentralized model fo…
Secure Linear MDS Coded Matrix Inversion
Neophytos Charalambides, Mert Pilanci, Alfred Hero
A cumbersome operation in many scientific fields, is inverting large full-rank matrices. In this paper, we propose a coded computing approach for recovering matrix inverse approxim…
Orthonormal Sketches for Secure Coded Regression
Neophytos Charalambides, Hessam Mahdavifar, Mert Pilanci +1
In this work, we propose a method for speeding up linear regression distributively, while ensuring security. We leverage randomized sketching techniques, and improve straggler resi…
Minimax Optimal Quantization of Linear Models: Information-Theoretic Limits and Efficient Algorithms
Rajarshi Saha, Mert Pilanci, Andrea J. Goldsmith
High-dimensional models often have a large memory footprint and must be quantized after training before being deployed on resource-constrained edge devices for inference tasks. In…
Computational Polarization: An Information-theoretic Method for Resilient Computing
Mert Pilanci
We introduce an error resilient distributed computing method based on an extension of the channel polarization phenomenon to distributed algorithms. The method leverages an algorit…
Adaptive and Oblivious Randomized Subspace Methods for High-Dimensional Optimization: Sharp Analysis and Lower Bounds
Jonathan Lacotte, Mert Pilanci
We propose novel randomized optimization methods for high-dimensional convex problems based on restrictions of variables to random subspaces. We consider oblivious and data-adaptiv…
Approximate Weighted Coded Matrix Multiplication
Neophytos Charalambides, Mert Pilanci, Alfred Hero
One of the most common, but at the same time expensive operations in linear algebra, is multiplying two matrices and . With the rapid development of machine learning and inc…
Weighted Gradient Coding with Leverage Score Sampling
Neophytos Charalambides, Mert Pilanci, Alfred O. Hero
A major hurdle in machine learning is scalability to massive datasets. Approaches to overcome this hurdle include compression of the data matrix and distributing the computations.…
Straggler Resilient Serverless Computing Based on Polar Codes
Burak Bartan, Mert Pilanci
We propose a serverless computing mechanism for distributed computation based on polar codes. Serverless computing is an emerging cloud based computation model that lets users run…
Randomized Sketches of Convex Programs with Sharp Guarantees
Mert Pilanci, Martin J. Wainwright
Random projection (RP) is a classical technique for reducing storage and computational costs. We analyze RP-based approximations of convex programs, in which the original optimizat…