Results 11 to 20 of about 1,470,777 (390)
ProPILE: Probing Privacy Leakage in Large Language Models [PDF]
Neural Information Processing Systems, 2023 The rapid advancement and widespread use of large language models (LLMs) have raised significant concerns regarding the potential leakage of personally identifiable information (PII).
Siwon Kim +5 more
semanticscholar +1 more source
Leakage and the Reproducibility Crisis in ML-based Science [PDF]
arXiv.org, 2022 The use of machine learning (ML) methods for prediction and forecasting has become widespread across the quantitative sciences. However, there are many known methodological pitfalls, including data leakage, in ML-based science.
Sayash Kapoor, Arvind Narayanan
semanticscholar +1 more source
Removing leakage-induced correlated errors in superconducting quantum error correction [PDF]
Nature Communications, 2021 Quantum computing can become scalable through error correction, but logical error rates only decrease with system size when physical errors are sufficiently uncorrelated.
M. McEwen +50 more
semanticscholar +1 more source
Overcoming leakage in quantum error correction [PDF]
Nature Physics, 2022 The leakage of quantum information out of the two computational states of a qubit into other energy states represents a major challenge for quantum error correction.
K. Miao +116 more
semanticscholar +1 more source
Gradient-Leakage Resilient Federated Learning [PDF]
IEEE International Conference on Distributed Computing Systems, 2021 Federated learning(FL) is an emerging distributed learning paradigm with default client privacy because clients can keep sensitive data on their devices and only share local training parameter updates with the federated server.
Wenqi Wei +4 more
semanticscholar +1 more source
Exploiting Unintended Feature Leakage in Collaborative Learning [PDF]
IEEE Symposium on Security and Privacy, 2018 Collaborative machine learning and related techniques such as federated learning allow multiple participants, each with his own training dataset, to build a joint model by training locally and periodically exchanging model updates.
Luca Melis +3 more
semanticscholar +1 more source
Membership Leakage in Label-Only Exposures [PDF]
Conference on Computer and Communications Security, 2020 Machine learning (ML) has been widely adopted in various privacy-critical applications, e.g., face recognition and medical image analysis. However, recent research has shown that ML models are vulnerable to attacks against their training data. Membership
Zheng Li, Yang Zhang
semanticscholar +1 more source
Information Leakage in Embedding Models [PDF]
Conference on Computer and Communications Security, 2020 Embeddings are functions that map raw input data to low-dimensional vector representations, while preserving important semantic information about the inputs.
Congzheng Song, A. Raghunathan
semanticscholar +1 more source
Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning [PDF]
Conference on Computer and Communications Security, 2017 Deep Learning has recently become hugely popular in machine learning for its ability to solve end-to-end learning systems, in which the features and the classifiers are learned simultaneously, providing significant improvements in classification accuracy
B. Hitaj, G. Ateniese, F. Pérez-Cruz
semanticscholar +1 more source
Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning [PDF]
IEEE Conference on Computer Communications, 2018 Federated learning, i.e., a mobile edge computing framework for deep learning, is a recent advance in privacy-preserving machine learning, where the model is trained in a decentralized manner by the clients, i.e., data curators, preventing the server ...
Zhibo Wang +5 more
semanticscholar +1 more source