TU Berlin

Fachgebiet Algorithmische AlgebraPhilipp Reichenbach

Page Content

to Navigation

There is no English translation for this web page.

Wissenschaftliche Mitarbeiter

Philipp Reichenbach

Lupe

Anschrift
Technische Universität Berlin
Institut für Mathematik
Sekretariat MA 3-2
Straße des 17. Juni 136
10623 Berlin
 
Büro
Raum MA 319 (3. OG)
Institut für Mathematik
 
Persönliche Homepage:
page.math.tu-berlin.de/~reichenb/
 
ORCiD
0000-0002-5722-5505

Kontakt

Sekretariat
Beate Nießen
Raum MA 318
Tel.: +49 (0)30 314 - 25771

eMail
reichenb@math.tu-berlin.de 

Telefon
+49 (0)30 314-29280
Faxgerät
+49 (0)30 314 - 25839
Sprechstunde
nach Vereinbarung

Publikationen in der Arbeitsgruppe

Barriers for recent methods in geodesic optimization
Citation key FR-Barriers-for-recent-methods-in-geodesic-optimization
Author Cole Franks and Philipp Reichenbach
Year 2021
Month 02
Abstract We study a class of optimization problems including matrix scaling, matrix balancing, multidimensional array scaling, operator scaling, and tensor scaling that arise frequently in theory and in practice. Some of these problems, such as matrix and array scaling, are convex in the Euclidean sense, but others such as operator scaling and tensor scaling are \emphgeodesically convex on a different Riemannian manifold. Trust region methods, which include box-constrained Newton's method, are known to produce high precision solutions very quickly for matrix scaling and matrix balancing (Cohen et. al., FOCS 2017, Allen-Zhu et. al. FOCS 2017), and result in polynomial time algorithms for some geodesically convex problems like operator scaling (Garg et. al. STOC 2018, Bürgisser et. al. FOCS 2019). One is led to ask whether these guarantees also hold for multidimensional array scaling and tensor scaling. We show that this is not the case by exhibiting instances with exponential \emphdiameter bound: we construct polynomial-size instances of 3-dimensional array scaling and 3-tensor scaling whose approximate solutions all have doubly exponential condition number. Moreover, we study convex-geometric notions of complexity known as margin and gap, which are used to bound the running times of all existing optimization algorithms for such problems. We show that margin and gap are exponentially small for several problems including array scaling, tensor scaling and polynomial scaling. Our results suggest that it is impossible to prove polynomial running time bounds for tensor scaling based on diameter bounds alone. Therefore, our work motivates the search for analogues of more sophisticated algorithms, such as interior point methods, for geodesically convex optimization that do not rely on polynomial diameter bounds.
Link to publication Download Bibtex entry

Navigation

Quick Access

Schnellnavigation zur Seite über Nummerneingabe