Zobrazeno 1 - 10
of 73
pro vyhledávání: '"Yaroslavtsev, Grigory"'
Private optimization is a topic of major interest in machine learning, with differentially private stochastic gradient descent (DP-SGD) playing a key role in both theory and practice. Furthermore, DP-SGD is known to be a powerful tool in contexts bey
Externí odkaz:
http://arxiv.org/abs/2410.06878
Contrastive learning is a highly successful technique for learning representations of data from labeled tuples, specifying the distance relations within the tuple. We study the sample complexity of contrastive learning, i.e. the minimum number of lab
Externí odkaz:
http://arxiv.org/abs/2312.00379
Autor:
Avdiukhin, Dmitrii, Yaroslavtsev, Grigory, Vainstein, Danny, Fischer, Orr, Das, Sauman, Mirza, Faraz
We study the problem of learning a hierarchical tree representation of data from labeled samples, taken from an arbitrary (and possibly adversarial) distribution. Consider a collection of data tuples labeled according to their hierarchical structure.
Externí odkaz:
http://arxiv.org/abs/2302.04492
We give the first polynomial time algorithms for escaping from high-dimensional saddle points under a moderate number of constraints. Given gradient access to a smooth function $f \colon \mathbb R^d \to \mathbb R$ we show that (noisy) gradient descen
Externí odkaz:
http://arxiv.org/abs/2205.13753
Stochastic gradient descent (SGD) is a prevalent optimization technique for large-scale distributed machine learning. While SGD computation can be efficiently divided between multiple machines, communication typically becomes a bottleneck in the dist
Externí odkaz:
http://arxiv.org/abs/2105.10090
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence (2021), 35(10), 9055-9063
We initiate a comprehensive experimental study of objective-based hierarchical clustering methods on massive datasets consisting of deep embedding vectors from computer vision and NLP applications. This includes a large variety of image embedding (Im
Externí odkaz:
http://arxiv.org/abs/2012.08466
Autor:
Ahmadian, Sara, Chatziafratis, Vaggos, Epasto, Alessandro, Lee, Euiwoong, Mahdian, Mohammad, Makarychev, Konstantin, Yaroslavtsev, Grigory
Hierarchical Clustering is an unsupervised data analysis method which has been widely used for decades. Despite its popularity, it had an underdeveloped analytical foundation and to address this, Dasgupta recently introduced an optimization viewpoint
Externí odkaz:
http://arxiv.org/abs/1912.06983
Autor:
Yaroslavtsev, Grigory, Zhou, Samson
A function $f : \mathbb{F}_2^n \to \mathbb{R}$ is $s$-sparse if it has at most $s$ non-zero Fourier coefficients. Motivated by applications to fast sparse Fourier transforms over $\mathbb{F}_2^n$, we study efficient algorithms for the problem of appr
Externí odkaz:
http://arxiv.org/abs/1910.05686
The problem of selecting a small-size representative summary of a large dataset is a cornerstone of machine learning, optimization and data science. Motivated by applications to recommendation systems and other scenarios with query-limited access to
Externí odkaz:
http://arxiv.org/abs/1910.05646
Autor:
Yaroslavtsev, Grigory, Zhou, Samson
We study the problem of constructing a linear sketch of minimum dimension that allows approximation of a given real-valued function $f \colon \mathbb{F}_2^n \rightarrow \mathbb R$ with small expected squared error. We develop a general theory of line
Externí odkaz:
http://arxiv.org/abs/1907.00524