Timezone: »
Poster
Computational Optimal Transport: Complexity by Accelerated Gradient Descent Is Better Than by Sinkhorn's Algorithm
Pavel Dvurechenskii · Alexander Gasnikov · Alexey Kroshnin
We analyze two algorithms for approximating the general optimal transport (OT) distance between two discrete distributions of size $n$, up to accuracy $\varepsilon$. For the first algorithm, which is based on the celebrated Sinkhorn's algorithm, we prove the complexity bound $\widetilde{O}\left(\frac{n^2}{\varepsilon^2}\right)$ arithmetic operations ($\widetilde{O}$ hides polylogarithmic factors $(\ln n)^c$, $c>0$). For the second one, which is based on our novel Adaptive Primal-Dual Accelerated Gradient Descent (APDAGD) algorithm, we prove the complexity bound $\widetilde{O}\left(\min\left\{\frac{n^{9/4}}{\varepsilon}, \frac{n^{2}}{\varepsilon^2} \right\}\right)$ arithmetic operations. Both bounds have better dependence on $\varepsilon$ than the state-of-the-art result given by $\widetilde{O}\left(\frac{n^2}{\varepsilon^3}\right)$. Our second algorithm not only has better dependence on $\varepsilon$ in the complexity bound, but also is not specific to entropic regularization and can solve the OT problem with different regularizers.
Author Information
Pavel Dvurechenskii (Weierstrass Institute for Applied Analysis and Stochastics)
Graduated with honors from Moscow Institute of Physics and Technology. PhD on differential games in the same university. At the moment research associate in the area of optimization under inexact information in Berlin. Research interest include - algorithms for convex and non-convex large-scale optimization problems; - optimization under deterministic and stochastic inexact information; - randomized algorithms: random coordinate descent, random (derivative-free) directional search; - numerical aspects of Optimal Transport - Algorithms for saddle-point problems and variational inequalities
Alexander Gasnikov (Moscow Institute of Physics and Technology)
Alexey Kroshnin (Institute for Information Transmission Problems)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Oral: Computational Optimal Transport: Complexity by Accelerated Gradient Descent Is Better Than by Sinkhorn's Algorithm »
Wed. Jul 11th 02:40 -- 02:50 PM Room A9
More from the Same Authors
-
2023 : Kernel Mirror Prox and RKHS Gradient Flow for Mixed Functional Nash Equilibrium »
Pavel Dvurechenskii · Jia-Jie Zhu -
2023 : Kernel Mirror Prox and RKHS Gradient Flow for Mixed Functional Nash Equilibrium »
Pavel Dvurechenskii · Jia-Jie Zhu -
2023 Poster: High-Probability Bounds for Stochastic Optimization and Variational Inequalities: the Case of Unbounded Variance »
Abdurakhmon Sadiev · Marina Danilova · Eduard Gorbunov · Samuel Horváth · Gauthier Gidel · Pavel Dvurechenskii · Alexander Gasnikov · Peter Richtarik -
2023 Poster: Is Consensus Acceleration Possible in Decentralized Optimization over Slowly Time-Varying Networks? »
Dmitry Metelev · Alexander Rogozin · Dmitry Kovalev · Alexander Gasnikov -
2022 Poster: The power of first-order smooth optimization for black-box non-smooth problems »
Alexander Gasnikov · Anton Novitskii · Vasilii Novitskii · Farshed Abdukhakimov · Dmitry Kamzolov · Aleksandr Beznosikov · Martin Takac · Pavel Dvurechenskii · Bin Gu -
2022 Spotlight: The power of first-order smooth optimization for black-box non-smooth problems »
Alexander Gasnikov · Anton Novitskii · Vasilii Novitskii · Farshed Abdukhakimov · Dmitry Kamzolov · Aleksandr Beznosikov · Martin Takac · Pavel Dvurechenskii · Bin Gu -
2021 Poster: ADOM: Accelerated Decentralized Optimization Method for Time-Varying Networks »
Dmitry Kovalev · Egor Shulgin · Peter Richtarik · Alexander Rogozin · Alexander Gasnikov -
2021 Spotlight: ADOM: Accelerated Decentralized Optimization Method for Time-Varying Networks »
Dmitry Kovalev · Egor Shulgin · Peter Richtarik · Alexander Rogozin · Alexander Gasnikov -
2021 Poster: On a Combination of Alternating Minimization and Nesterov's Momentum »
Sergey Guminov · Pavel Dvurechenskii · Nazarii Tupitsa · Alexander Gasnikov -
2021 Spotlight: On a Combination of Alternating Minimization and Nesterov's Momentum »
Sergey Guminov · Pavel Dvurechenskii · Nazarii Tupitsa · Alexander Gasnikov -
2021 Poster: Newton Method over Networks is Fast up to the Statistical Precision »
Amir Daneshmand · Gesualdo Scutari · Pavel Dvurechenskii · Alexander Gasnikov -
2021 Spotlight: Newton Method over Networks is Fast up to the Statistical Precision »
Amir Daneshmand · Gesualdo Scutari · Pavel Dvurechenskii · Alexander Gasnikov -
2020 Poster: Self-Concordant Analysis of Frank-Wolfe Algorithms »
Pavel Dvurechenskii · Petr Ostroukhov · Kamil Safin · Shimrit Shtern · Mathias Staudigl -
2019 Poster: On the Complexity of Approximating Wasserstein Barycenters »
Alexey Kroshnin · Nazarii Tupitsa · Darina Dvinskikh · Pavel Dvurechenskii · Alexander Gasnikov · Cesar Uribe -
2019 Oral: On the Complexity of Approximating Wasserstein Barycenters »
Alexey Kroshnin · Nazarii Tupitsa · Darina Dvinskikh · Pavel Dvurechenskii · Alexander Gasnikov · Cesar Uribe