Zobrazeno 1 - 10
of 489
pro vyhledávání: '"MEHROTRA, SHARAD"'
Graph Structure Prompt Learning: A Novel Methodology to Improve Performance of Graph Neural Networks
Graph neural networks (GNNs) are widely applied in graph data modeling. However, existing GNNs are often trained in a task-driven manner that fails to fully capture the intrinsic nature of the graph structure, resulting in sub-optimal node and graph
Externí odkaz:
http://arxiv.org/abs/2407.11361
Despite the Graph Neural Networks' (GNNs) proficiency in analyzing graph data, achieving high-accuracy and interpretable predictions remains challenging. Existing GNN interpreters typically provide post-hoc explanations disjointed from GNNs' predicti
Externí odkaz:
http://arxiv.org/abs/2407.11358
This paper studies privacy in the context of complex decision support queries composed of multiple conditions on different aggregate statistics combined using disjunction and conjunction operators. Utility requirements for such queries necessitate th
Externí odkaz:
http://arxiv.org/abs/2406.15655
This paper addresses volume leakage (i.e., leakage of the number of records in the answer set) when processing keyword queries in encrypted key-value (KV) datasets. Volume leakage, coupled with prior knowledge about data distribution and/or previousl
Externí odkaz:
http://arxiv.org/abs/2310.12491
We present a novel inference scheme, self-speculative decoding, for accelerating Large Language Models (LLMs) without the need for an auxiliary model. This approach is characterized by a two-stage process: drafting and verification. The drafting stag
Externí odkaz:
http://arxiv.org/abs/2309.08168
Autor:
Chakraborty, Vishal, Ann-Elvy, Stacy, Mehrotra, Sharad, Nawab, Faisal, Sadoghi, Mohammad, Sharma, Shantanu, Venkatsubhramanian, Nalini, Saeed, Farhan
Data regulations, such as GDPR, are increasingly being adopted globally to protect against unsafe data management practices. Such regulations are, often ambiguous (with multiple valid interpretations) when it comes to defining the expected dynamic be
Externí odkaz:
http://arxiv.org/abs/2308.07501
We propose Transactional Edge (TransEdge), a distributed transaction processing system for untrusted environments such as edge computing systems. What distinguishes TransEdge is its focus on efficient support for read-only transactions. TransEdge all
Externí odkaz:
http://arxiv.org/abs/2302.08019
Autor:
Elkordy, Ahmed Roushdy, Ezzeldin, Yahya H., Han, Shanshan, Sharma, Shantanu, He, Chaoyang, Mehrotra, Sharad, Avestimehr, Salman
Publikováno v:
APSIPA Transactions on Signal and Information Processing, Volume 12, Issue 1, 2023
Federated analytics (FA) is a privacy-preserving framework for computing data analytics over multiple remote parties (e.g., mobile devices) or silo-ed institutional entities (e.g., hospitals, banks) without sharing the data among parties. Motivated b
Externí odkaz:
http://arxiv.org/abs/2302.01326
Simply restricting the computation to non-sensitive part of the data may lead to inferences on sensitive data through data dependencies. Inference control from data dependencies has been studied in the prior work. However, existing solutions either d
Externí odkaz:
http://arxiv.org/abs/2207.08757
Autor:
Lin, Yiming, Mehrotra, Sharad
Missing values widely exist in real-world data sets, and failure to clean the missing data may result in the poor quality of answers to queries. \yiming{Traditionally, missing value imputation has been studied as an offline process as part of prepari
Externí odkaz:
http://arxiv.org/abs/2204.00108