Zobrazeno 1 - 10
of 382
pro vyhledávání: '"Khisti, Ashish"'
This paper investigates a novel lossy compression framework operating under logarithmic loss, designed to handle situations where the reconstruction distribution diverges from the source distribution. This framework is especially relevant for applica
Externí odkaz:
http://arxiv.org/abs/2410.21666
Autor:
Khisti, Ashish, Ebrahimi, M. Reza, Dbouk, Hassan, Behboodi, Arash, Memisevic, Roland, Louizos, Christos
We consider multi-draft speculative sampling, where the proposal sequences are sampled independently from different draft models. At each step, a token-level draft selection scheme takes a list of valid tokens as input and produces an output token wh
Externí odkaz:
http://arxiv.org/abs/2410.18234
Parameter-Efficient Fine-Tuning (PEFT) has risen as an innovative training strategy that updates only a select few model parameters, significantly lowering both computational and memory demands. PEFT also helps to decrease data transfer in federated
Externí odkaz:
http://arxiv.org/abs/2409.02346
Autor:
Qian, Jingjing, Salehkalaibar, Sadaf, Chen, Jun, Khisti, Ashish, Yu, Wei, Shi, Wuxian, Ge, Yiqun, Tong, Wen
This paper studies the rate-distortion-perception (RDP) tradeoff for a Gaussian vector source coding problem where the goal is to compress the multi-component source subject to distortion and perception constraints. The purpose of imposing a percepti
Externí odkaz:
http://arxiv.org/abs/2406.18008
This survey articles focuses on emerging connections between the fields of machine learning and data compression. While fundamental limits of classical (lossy) data compression are established using rate-distortion theory, the connections to machine
Externí odkaz:
http://arxiv.org/abs/2406.10036
We propose a secure inference protocol for a distributed setting involving a single server node and multiple client nodes. We assume that the observed data vector is partitioned across multiple client nodes while the deep learning model is located at
Externí odkaz:
http://arxiv.org/abs/2405.03775
Autor:
Chen, Shuangyi, Khisti, Ashish
In the context of prediction-as-a-service, concerns about the privacy of the data and the model have been brought up and tackled via secure inference protocols. These protocols are built up by using single or multiple cryptographic tools designed und
Externí odkaz:
http://arxiv.org/abs/2404.16232
This paper investigates adaptive streaming codes over a three-node relayed network. In this setting, a source transmits a sequence of message packets through a relay under a delay constraint of $T$ time slots per packet. The source-to-relay and relay
Externí odkaz:
http://arxiv.org/abs/2401.15056
We study the rate-distortion-perception (RDP) tradeoff for a memoryless source model in the asymptotic limit of large block-lengths. Our perception measure is based on a divergence between the distributions of the source and reconstruction sequences
Externí odkaz:
http://arxiv.org/abs/2401.12207
We propose two extensions to existing importance sampling based methods for lossy compression. First, we introduce an importance sampling based compression scheme that is a variant of ordered random coding (Theis and Ahmed, 2022) and is amenable to d
Externí odkaz:
http://arxiv.org/abs/2401.02609