Zobrazeno 1 - 10
of 46
pro vyhledávání: '"Drori, Yoel"'
We introduce a novel dynamic learning-rate scheduling scheme grounded in theory with the goal of simplifying the manual and time-consuming tuning of schedules in practice. Our approach is based on estimating the locally-optimal stepsize, guaranteeing
Externí odkaz:
http://arxiv.org/abs/2311.13877
Autor:
Tang, Jiaxi, Drori, Yoel, Chang, Daryl, Sathiamoorthy, Maheswaran, Gilmer, Justin, Wei, Li, Yi, Xinyang, Hong, Lichan, Chi, Ed H.
Recommender systems play an important role in many content platforms. While most recommendation research is dedicated to designing better models to improve user experience, we found that research on stabilizing the training for such models is severel
Externí odkaz:
http://arxiv.org/abs/2302.09178
We consider stochastic optimization with delayed gradients where, at each time step $t$, the algorithm makes an update using a stale stochastic gradient from step $t - d_t$ for some arbitrary delay $d_t$. This setting abstracts asynchronous distribut
Externí odkaz:
http://arxiv.org/abs/2106.11879
Autor:
Drori, Yoel, Taylor, Adrien
We construct a family of functions suitable for establishing lower bounds on the oracle complexity of first-order minimization of smooth strongly-convex functions. Based on this construction, we derive new lower bounds on the complexity of strongly-c
Externí odkaz:
http://arxiv.org/abs/2101.09740
Autor:
Taylor, Adrien, Drori, Yoel
We present an optimal gradient method for smooth strongly convex optimization. The method is optimal in the sense that its worst-case bound on the distance to an optimal point exactly matches the lower bound on the oracle complexity for the class of
Externí odkaz:
http://arxiv.org/abs/2101.09741
Autor:
Drori, Yoel, Shamir, Ohad
We study the iteration complexity of stochastic gradient descent (SGD) for minimizing the gradient norm of smooth, possibly nonconvex functions. We provide several results, implying that the $\mathcal{O}(\epsilon^{-4})$ upper bound of Ghadimi and Lan
Externí odkaz:
http://arxiv.org/abs/1910.01845
We study a variant of domain adaptation for named-entity recognition where multiple, heterogeneously tagged training sets are available. Furthermore, the test tag-set is not identical to any individual training tag-set. Yet, the relations between all
Externí odkaz:
http://arxiv.org/abs/1905.09135
Autor:
Drori, Yoel
We consider the class of smooth convex functions defined over an open convex set. We show that this class is essentially different than the class of smooth convex functions defined over the entire linear space by exhibiting a function that belongs to
Externí odkaz:
http://arxiv.org/abs/1812.02419
Autor:
Drori, Yoel, Taylor, Adrien B.
We describe a novel constructive technique for devising efficient first-order methods for a wide range of large-scale convex minimization settings, including smooth, non-smooth, and strongly convex minimization. The technique builds upon a certain va
Externí odkaz:
http://arxiv.org/abs/1803.05676
Autor:
Drori, Yoel
We obtain a new lower bound on the information-based complexity of first-order minimization of smooth and convex functions. We show that the bound matches the worst-case performance of the recently introduced Optimized Gradient Method, thereby establ
Externí odkaz:
http://arxiv.org/abs/1606.01424