Authors : Pustelnik, Nelly (Author of the conference)
CIRM (Publisher )
Abstract :
During the last 20 years, imaging sciences, including inverse problems, segmentation or classification, has known two major revolutions: (i) sparsity and proximal algorithms and (ii) deep learning and stochastic optimization. This course proposes to illustrate these major advances in the context of imaging problems that can be formulated as the minimization of an objective function and to highlight the evolution of these objective functions jointly with optimization advances.
Since 2003, convex optimization has become the main thrust behind significant advances in signal processing, image processing and machine learning. The increasingly complex variational formulations encountered in these areas which may involve a sum of several, possibly non-smooth, convex terms, together with the large sizes of the problems at hand make the use of standard optimization methods such as those based on subgradient descent techniques intractable computationally. Since their introduction in the signal processing arena, splitting techniques have emerged as a central tool to circumvent these roadblocks: they operate by breaking down the problem into individual components that can be activated individually in the solution algorithm. In the past decade, numerous convex optimization algorithms based on splitting techniques have been proposed or rediscovered in an attempt to efficiently deal with such problems. We will provide the basic building blocks for major proximal algorithm strategies and their recent advances in nonconvex and stochastic optimization. Behind non-smooth functions, there is the concept of sparsity which is central in the contributions in inverse problems and compressed sensing. This concept will be described as well as the objective functions relying on it, going from Mumford-Shah model to sparse SVM. Ten years after the start of proximal revolution, deep learning has started to provide a new framework for solving imaging problems going from agnostic techniques to models combining deep learning with standard regularized formulation. The main encountered objective functions as well as the associated algorithmic strategies will be discussed.
1/ Introduction
2/ Optimization: basics
3/ Subdifferential and proximity operator
4/ First order schemes (gradient descent, proximal point algorithm, forward-backward splitting, Peaceman-Rachford splitting, Douglas-Rachford splitting): weak and linear convergence.
5/ Conjugate, duality, proximal primal-dual algorithms
6/ Unfolded algorithms
7/ Acceleration, non-convex optimization, stochastic optimization
MSC Codes :
49N45
- Inverse problems in calculus of variations
94A08
- Image processing (compression, reconstruction, etc.)
|
Event Title : Mathematics, Signal Processing and Learning / Mathématiques, traitement du signal et apprentissage Event Organizers : Anthoine, Sandrine ; Chaux, Caroline ; Mélot, Clothilde ; Richard, Frédéric Dates : 25/01/2021 - 29/01/2021
Event Year : 2021
Event URL : https://conferences.cirm-math.fr/2472.html
DOI : 10.24350/CIRM.V.19706303
Cite this video as:
Pustelnik, Nelly (2021). Optimization - lecture 2. CIRM. Audiovisual resource. doi:10.24350/CIRM.V.19706303
URI : http://dx.doi.org/10.24350/CIRM.V.19706303
|
See Also
-
[Multi angle]
Teasing poster: mathematics, signal processing and learning
/ Author of the conference Antonsanti, Pierre-Louis ; Author of the conference Belotto Da Silva, André ; Author of the conference Cano, Cyril ; Author of the conference Cohen, Jeremy ; Author of the conference Doz, Cyprien ; Author of the conference Lazzaretti, Marta ; Author of the conference Pilavci, Yusuf Yigit ; Author of the conference Rodriguez, Willy ; Author of the conference Stergiopoulou, Vasiliki ; Author of the conference Kaloga, Yacouba ; Author of the conference Safaa, Al-Ali.
-
[Virtualconference]
Optimization - lecture 4
/ Author of the conference Pustelnik, Nelly.
-
[Virtualconference]
Optimization - lecture 3
/ Author of the conference Pustelnik, Nelly.
-
[Virtualconference]
Optimization - lecture 1
/ Author of the conference Pustelnik, Nelly.
-
[Multi angle]
One signal processing view on deep learning - lecture 2
/ Author of the conference Oyallon, Edouard.
-
[Multi angle]
One signal processing view on deep learning - lecture 1
/ Author of the conference Oyallon, Edouard.
-
[Virtualconference]
Signal processing tutorial - part 2
/ Author of the conference Oudre, Laurent.
-
[Virtualconference]
Signal processing tutorial - part 1
/ Author of the conference Oudre, Laurent.
-
[Virtualconference]
Reinforcement learning - lecture 4
/ Author of the conference Lazaric, Allesandro.
-
[Virtualconference]
Reinforcement learning - lecture 3
/ Author of the conference Lazaric, Allesandro.
-
[Virtualconference]
Reinforcement learning - lecture 2
/ Author of the conference Lazaric, Allesandro.
-
[Virtualconference]
Reinforcement learning - lecture 1
/ Author of the conference Lazaric, Allesandro.
-
[Multi angle]
Basics in machine learning - practical session 2
/ Author of the conference Clausel, Marianne.
-
[Multi angle]
Basics in machine learning - practical session 1
/ Author of the conference Clausel, Marianne.
-
[Multi angle]
Basics in machine learning - lecture 2
/ Author of the conference Clausel, Marianne.
-
[Multi angle]
Basics in machine learning - lecture 1
/ Author of the conference Clausel, Marianne.
Bibliography