Zobrazeno 1 - 10
of 77
pro vyhledávání: '"Krichene, Walid"'
Autor:
Chua, Lynn, Cui, Qiliang, Ghazi, Badih, Harrison, Charlie, Kamath, Pritish, Krichene, Walid, Kumar, Ravi, Manurangsi, Pasin, Narra, Krishna Giri, Sinha, Amer, Varadarajan, Avinash, Zhang, Chiyuan
Motivated by problems arising in digital advertising, we introduce the task of training differentially private (DP) machine learning models with semi-sensitive features. In this setting, a subset of the features is known to the attacker (and thus nee
Externí odkaz:
http://arxiv.org/abs/2401.15246
Autor:
Krichene, Walid, Mayoraz, Nicolas, Rendle, Steffen, Song, Shuang, Thakurta, Abhradeep, Zhang, Li
We study a class of private learning problems in which the data is a join of private and public features. This is often the case in private personalization tasks such as recommendation or ad prediction, in which features related to individuals are se
Externí odkaz:
http://arxiv.org/abs/2310.15454
We consider the problem of training private recommendation models with access to public item features. Training with Differential Privacy (DP) offers strong privacy guarantees, at the expense of loss in recommendation quality. We show that incorporat
Externí odkaz:
http://arxiv.org/abs/2309.11516
Autor:
Krichene, Walid, Jain, Prateek, Song, Shuang, Sundararajan, Mukund, Thakurta, Abhradeep, Zhang, Li
We study the problem of multi-task learning under user-level differential privacy, in which $n$ users contribute data to $m$ tasks, each involving a subset of users. One important aspect of the problem, that can significantly impact quality, is the d
Externí odkaz:
http://arxiv.org/abs/2302.07975
Leveraging transfer learning has recently been shown to be an effective strategy for training large models with Differential Privacy (DP). Moreover, somewhat surprisingly, recent works have found that privately training just the last layer of a pre-t
Externí odkaz:
http://arxiv.org/abs/2211.13403
Autor:
Sundararajan, Mukund, Krichene, Walid
Machine learning is pervasive. It powers recommender systems such as Spotify, Instagram and YouTube, and health-care systems via models that predict sleep patterns, or the risk of disease. Individuals contribute data to these models and benefit from
Externí odkaz:
http://arxiv.org/abs/2202.09480
We present ALX, an open-source library for distributed matrix factorization using Alternating Least Squares, written in JAX. Our design allows for efficient use of the TPU architecture and scales well to matrix factorization problems of O(B) rows/col
Externí odkaz:
http://arxiv.org/abs/2112.02194
iALS is a popular algorithm for learning matrix factorization models from implicit feedback with alternating least squares. This algorithm was invented over a decade ago but still shows competitive quality compared to recent approaches like VAE, EASE
Externí odkaz:
http://arxiv.org/abs/2110.14044
Matrix factorization learned by implicit alternating least squares (iALS) is a popular baseline in recommender system research publications. iALS is known to be one of the most computationally efficient and scalable collaborative filtering methods. H
Externí odkaz:
http://arxiv.org/abs/2110.14037
Autor:
Chien, Steve, Jain, Prateek, Krichene, Walid, Rendle, Steffen, Song, Shuang, Thakurta, Abhradeep, Zhang, Li
We study the problem of differentially private (DP) matrix completion under user-level privacy. We design a joint differentially private variant of the popular Alternating-Least-Squares (ALS) method that achieves: i) (nearly) optimal sample complexit
Externí odkaz:
http://arxiv.org/abs/2107.09802