Continuous Optimization Seminar - Ryan KinnearExport this event to calendar

Wednesday, November 1, 2017 — 4:00 PM EDT

Title: A Stochastic Gradient Method with an Exponential Convergence Rate for Finite Training Sets

Speaker: Ryan Kinnear
Affiliation: University of Waterloo
Room: MC 5479

Abstract:

We will be discussing the paper (having the same title) by Roux, Schmidt, and Bach.  The authors propose a new stochastic gradient method for optimizing the sum of
 a finite set of smooth functions, where the sum is strongly convex.
 While standard stochastic gradient methods
 converge at sublinear rates for this problem, the proposed method incorporates a memory of previous gradient values in order to achieve a linear convergence 
rate. In a machine learning context, numerical experiments indicate that the new algorithm can dramatically outperform standard
 algorithms, both in terms of optimizing the training error and reducing the test error quickly.

Location 
MC - Mathematics & Computer Building
5479
200 University Avenue West

Waterloo, ON N2L 3G1
Canada

S M T W T F S
28
29
30
31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1
  1. 2023 (59)
    1. June (7)
    2. May (12)
    3. April (5)
    4. March (17)
    5. February (10)
    6. January (8)
  2. 2022 (150)
    1. December (8)
    2. November (18)
    3. October (15)
    4. September (11)
    5. August (2)
    6. July (17)
    7. June (17)
    8. May (10)
    9. April (12)
    10. March (18)
    11. February (10)
    12. January (13)
  3. 2021 (103)
  4. 2020 (119)
  5. 2019 (167)
  6. 2018 (136)
  7. 2017 (103)
  8. 2016 (137)
  9. 2015 (136)
  10. 2014 (88)
  11. 2013 (48)
  12. 2012 (39)
  13. 2011 (36)
  14. 2010 (40)
  15. 2009 (40)
  16. 2008 (39)
  17. 2007 (15)