Talks

Details view List view

Deep Learning

למידה עמוקה - המהפיכה שתשנה את חיינו
November 22, 2023
Zooom Meeting

למידה עמוקה (deep learning)  הוא תחום שישנה את חיינו – תיקון – הוא כבר משנה את חיינו. בהרצאה זו שניתנה לסטודנטים בטכניון ולמתעניינים אחרים אני מספר את סיפורו המרתק של התחום הזה ואת התהפוכות שהוא עבר בששים השנים האחרונות. הרצאה זו נועדה לקהל הרחב ולא דורשת כל ידע מוקדם.

קישור זה מפנה לסרט וידאו של הרצאה זו שהוקלט באולפן.

קישור זה מפנה לסרט וידאו של הקלטת מפגש הזום (איכות האודיו נומוכה יותר) בו העברתי את אותה הרצאה.

The New Era of Image Denoising - The Deep Learning Revolution and Beyond
April 18, 2022
Lanzhou University, Lanzhou, P.R. China

Part A: Image denoising – removal of white additive Gaussian noise from an image – is one of the oldest and most studied problems in image processing. An extensive work over several decades has led to thousands of papers on this subject, and to many well-performing algorithms for this task. As expected, the era of deep learning has brought yet another revolution to this subfield, and took the lead in today’s ability for noise suppression in images. All this progress has led some researchers to believe that “Denoising Is Dead”, in the sense that all that can be achieved is already done. Part A of this talk we will introduce the above evolution of this field, and highlight the tension that exists between classical approaches and modern AI alternatives.

Part B: Part B of this talk will focus on recently discovered abilities and vulnerabilities of image denoisers. In a nut-shell, we expose the possibility of using image denoisers for serving other problems, such as regularizing general inverse problems and serving as the engine for image synthesis. We also unveil the (strange?) idea that denoising (and other inverse problems) might not have a unique solution, as common algorithms would have you believe. Instead, we will describe constructive ways to produce randomized and diverse high perceptual quality results for inverse problems.

This was given as a plenary talk in the third international workshop on matrix computations, commemorating the 90th birthday of Gene Golub.

Image Denoising - Not What You Think
September 10, 2021
Invited talk- Berkeley, Rice, IMVC

Image denoising – removal of white additive Gaussian noise from an image – is one of the oldest and most studied problems in image processing. An extensive work over several decades has led to thousands of papers on this subject, and to many well-performing algorithms for this task. As expected, the era of deep learning has brought yet another revolution to this subfield, and took the lead in today’s ability for noise suppression in images. All this progress has led some researchers to believe that “denoising is dead”, in the sense that all that can be achieved is already done.

Exciting as all this story might be, this talk IS NOT ABOUT IT!

Our story focuses on recently discovered abilities and vulnerabilities of image denoisers. In a nut-shell, we expose the possibility of using image denoisers for serving other problems, such as regularizing general inverse problems and serving as the engine for image synthesis. We also unveil the (strange?) idea that denoising (and other inverse problems) might not have a unique solution, as common algorithms would have you believe. Instead, we will describe constructive ways to produce randomized and diverse high perceptual quality results for inverse problems.

A recording of this talk can be found HERE.

This talk was also given in the TCE-MLIS event on February 24th. Here is a recording of this talk (in Hebrew!)

 

Image Denoising - Not What You Think
July 13, 2021.
IEEE Statistical Signal Processing Workshop 2021 - Rio De Janeiro (Virtual) (Keynote Talk)

Image denoising – removal of white additive Gaussian noise from an image – is one of the oldest and most studied problems in image processing. An extensive work over several decades has led to thousands of papers on this subject, and to many well-performing algorithms for this task. As expected, the era of deep learning has brought yet another revolution to this subfield, and took the lead in today’s ability for noise suppression in images. All this progress has led some researchers to believe that “denoising is dead”, in the sense that all that can be achieved is already done.

Exciting as all this story might be, this talk IS NOT ABOUT it!

Our story focuses on recently discovered abilities and vulnerabilities of image denoisers. In a nut-shell, we expose the possibility of using image denoisers for serving other problems, such as regularizing general inverse problems and serving as the engine for image synthesis. We also unveil the (strange?) idea that denoising might not have a unique solution, as common algorithms would have you believe. Instead, we’ll describe constructive ways to produce randomized and diverse high perceptual quality denoising results.

A shorter version of this talk was given in June 17th as an Invited Talk in a conference on AI organized by RAFAEL.
למידה עמוקה - המהפיכה שתשנה את חיינו
March 22, 2021
צה"ל

למידה עמוקה (deep learning)  הוא תחום שישנה את חיינו – תיקון – הוא כבר משנה את חיינו. בהרצאה זו שניתנה לגורמי צה”ל, אני מספר את סיפורו המרתק של התחום הזה ואת התהפוכות שהוא עבר בששים השנים האחרונות. הרצאה זו נועדה לקהל הרחב ולא דורשת כל ידע מוקדם. זוהי גירסה ארוכה יותר של הרצאה דומה שניתנה ב- 2019 במסיבת היובל לפקולטה למדעי המחשב.

Design of Deep Learning Architectures
February 4, 2020.
Google Mountain-View - Computational Imaging Workshop (Keynote Talk)

How do we choose a network architecture in deep-learning solutions? By copying existing networks or guessing new ones, and sometimes by applying various small modifications to them via trial and error. This non-elegant and brute-force strategy has proven itself useful for a wide variety of imaging tasks. However, it comes with a painful cost – our networks tend to be quite heavy and cumbersome. Could we do better? In this talk we would like to propose a different point of view towards this important question, by advocating the following two rules: (i) Rather than “guessing” architectures, we should rely on classic signal and image processing concepts and algorithms, and turn these to networks to be learned in a supervised manner. More specifically, (ii) Sparse representation modeling is key in many (if not all) of the successful architectures that we are using. I will demonstrate these claims by presenting three recent image denoising networks that are light-weight and yet quite effective, as they follow the above guidelines.

Joint work with Peyman Milanfar.
Sparse Modeling of Data and its Relation to Deep Learning
November 1, 2019.
Princeton - DeepMath Conference (Invited Talk)

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model and its features, and then turn to describe two special cases of it – the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. Alongside this main message of bringing a theoretical backbone to deep-learning, another central message that will accompany us throughout the talk: Generative models for describing data sources enable a systematic way to design algorithms, while also providing a complete mechanism for a theoretical analysis of these algorithms’ performance. This talk is meant for newcomers to this field – no prior knowledge on sparse approximation is assumed.

This talk was also given in CAMSAP 2019 in Guadeloupe on December 17th 2019.
למידה עמוקה - המהפיכה שתשנה את חיינו
October 29, 2019.
יובל ה-50 לפקולטה למדעי המחשב בטכניון

למידה עמוקה (deep learning)  הוא תחום שישנה את חיינו – תיקון – הוא כבר משנה את חיינו. בהרצאה קצרה זו שניתנה ביובל ה-50 לפקולטה למדעי המחשב, אני מספר את סיפורו המרתק של התחום הזה ואת התהפוכות שהוא עבר בששים השנים האחרונות. הרצאה זו נועדה לקהל הרחב ולא דורשת כל ידע מוקדם.

Sparse Modelling of Data and its Relation to Deep Learning
June 27, 2019.
ETH - FIM - Institute for Mathematical Research: Series of Lectures on Waves and Imaging (III)

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model and its features, and then turn to describe two special cases of it – the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. Alongside this main message of bringing a theoretical backbone to deep-learning, another central message that will accompany us throughout the talk: Generative models for describing data sources enable a systematic way to design algorithms, while also providing a complete mechanism for a theoretical analysis of these algorithms’ performance. This talk is meant for newcomers to this field – no prior knowledge on sparse approximation is assumed.

This is a KEYNOTE talk in this event.
Sparse Modeling and Deep Learning
January 9, 2019.
QBI (Quantitative BioImaging Conference) 2019, Rennes, France

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model and its features, and then turn to describe two special cases of it – the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. Alongside this main message of bringing a theoretical backbone to deep-learning, another central message that will accompany us throughout the talk: Generative models for describing data sources enable a systematic way to design algorithms, while also providing a complete mechanism for a theoretical analysis of these algorithms’ performance. This talk is meant for newcomers to this field – no prior knowledge on sparse approximation is assumed.

This is a KEYNOTE talk in this event.
Sparse Modelling of Data and its Relation to Deep Learning
November 27, 2018.
EUVIP 2018, Tampere, Finland

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model, and then turn to describe two special cases of it – the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. This talk is meant for newcomers to these fields – no prior knowledge on sparse approximation is assumed.

This is a KEYNOTE talk in EUVIP 2018.
Sparse Modeling and Deep Learning
July 14, 2018.
ICML 2018, Stockholm Sweden

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk, we describe a special case of this model— the multi-layered convolutional sparse coding (ML-CSC) construction. As we will carefully show, ML-CSC provides a solid theoretical foundation to the field of deep learning, explaining the used architectures, their performance limits, and prospects for future alternatives.

This is an invited talk in an ICML workshop, titled "The theory of deep learning", and organized by Rene Vidal, Joan Bruna and Raja Giryes. The same talk has been given in a symposium on deep-learning in ICSEE in Eilat on December 13th 2018.
Sparse Modelling in Image Processing and Deep Learning
July 9, 2018.
THE 10th IEEE Sensor Array and Multichannel (SAM) Signal Processing Workshop, SHEFFIELD UK

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model and its features, and then turn to describe two special cases of it – the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. Alongside this main message of bringing a theoretical backbone to deep-learning, another central message that will accompany us throughout the talk: Generative models for describing data sources enable a systematic way to design algorithms, while also providing a complete mechanism for a theoretical analysis of these algorithms’ performance. This talk is meant for newcomers to this field – no prior knowledge on sparse approximation is assumed.

This is a PLENARY talk that was given in SAM 2018, in Sheffield, UK. This is a joint work with Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
Sparse Modeling in Image Processing and Deep Learning
March 6, 2018
IMVC 2018, Tel-Aviv, Israel

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model, and then turn to describe two special cases of it — the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to … deep-learning. This talk is meant for newcomers to these fields – no prior knowledge on sparse approximation is assumed.

This is a KEYNOTE talk that was given in IMVC 2018, in Tel-Aviv. This talk was also given in a seminar in the Hebrew University of Jerusalm on May 21st 2018. This is a joint work with Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
Sparse Modeling in Image Processing and Deep Learning
5-9 February 2018
IPAM 2018, Los Angeles, USA

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we describe two special cases of this model — the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). We show that the projection of signals (a.k.a. pursuit) to the ML-CSC model leads to various deep convolutional neural network architectures. This connection brings a fresh view to CNN, as we are able to accompany the above by theoretical claims such as uniqueness of the representations throughout the network, and their stable estimation, all guaranteed under simple local sparsity conditions. The ‘take-home-message’ from this talk is this: The ML-CSC model can serve as the theoretical foundation to deep-learning.

This is an invited talk that was given in IPAM (UCLA) during the "New Deep Learning Techniques" program during February 5-9, 2018, in Los Angeles, USA. This is a joint work with Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
Sparse Modeling in Image Processing and Deep Learning
18 of September 2017
ICIP 2017, in Beijing China

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk we start by presenting this model and its features, and then turn to describe two special cases of it: the convolutional sparse coding (CSC) and its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to the field of deep-learning. Alongside this main message of bringing a theoretical backbone to deep-learning, another central message that will accompany us throughout the talk: Generative models for describing data sources enable a systematic way to design algorithms, while also providing a complete mechanism for a theoretical analysis of these algorithms’ performance. This talk is meant for newcomers to this field – no prior knowledge on sparse approximation is assumed.

This is a KEYNOTE talk that was given in ICIP 2017, in Beijing China. This talk summarizes portions of the PhD work by my three PhD students, Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
From Sparse Representations to Deep Learning
4-8 September 2017
Summer School "Signal Processing meets Deep Learning" in Capri

Within the wide field of sparse approximation, convolutional sparse coding (CSC) has gained increasing attention in recent years. This model assumes a structured-dictionary built as a union of banded Circulant matrices. Most of the attention has been devoted to the practical side of CSC, proposing efficient algorithms for the pursuit problem, and identifying applications that benefit from this model. Interestingly, a systematic theoretical understanding of CSC seems to have been left aside, with the assumption that the existing classical results are sufficient. In this talk we start by presenting a novel analysis of the CSC model and its as- sociated pursuit. Our study is based on the observation that while being global, this model can be characterized and analyzed locally.

We show that uniqueness of the representation, its stability with respect to noise, and successful greedy or convex recovery are all guaranteed assuming that the underlying representation is locally sparse. These new results are much stronger and informative, compared to those obtained by deploying the classical sparse theory. Armed with these new insights, we proceed by proposing a multi-layer extension of this model, ML-CSC, in which signals are assumed to emerge from a cascade of CSC layers. This, in turn, is shown to be tightly connected to Convolutional Neural Networks (CNN), so much so that the forward-pass of the CNN is in fact the Thresholding pursuit serving the ML-CSC model. This connection brings a fresh view to CNN, as we are able to attribute to this architecture theoretical claims such as uniqueness of the representations throughout the network, and their stable estimation, all guaranteed under simple local sparsity conditions. Lastly, identifying the weaknesses in the above scheme, we propose an alternative to the forward-pass algorithm, which is both tightly connected to deconvolutional and recurrent neural networks, and has better theoretical guarantees.

This 3-hours talk was given at the Summer School "Signal Processing meets Deep Learning" in Capri. This talk summarizes portions of the PhD work by my three PhD students, Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
A Tale of Signal Modeling Evolution: SparseLand to CSC to CNN
June 1st, 2017
National University of Singapore

Within the wide field of sparse approximation, convolutional sparse coding (CSC) has gained increasing attention in recent years. This model assumes a structured-dictionary built as a union of banded Circulant matrices. Most of the attention has been devoted to the practical side of CSC, proposing efficient algorithms for the pursuit problem, and identifying applications that benefit from this model. Interestingly, a systematic theoretical understanding of CSC seems to have been left aside, with the assumption that the existing classical results are sufficient. In this talk we start by presenting a novel analysis of the CSC model and its as- sociated pursuit. Our study is based on the observation that while being global, this model can be characterized and analyzed locally. We show that uniqueness of the representation, its stability with respect to noise, and successful greedy or convex recovery are all guaranteed assuming that the underlying representation is locally sparse.

These new results are much stronger and informative, compared to those obtained by deploying the classical sparse theory. Armed with these new insights, we proceed by proposing a multi-layer extension of this model, ML-CSC, in which signals are assumed to emerge from a cascade of CSC layers. This, in turn, is shown to be tightly connected to Convolutional Neural Networks (CNN), so much so that the forward-pass of the CNN is in fact the Thresholding pursuit serving the ML-CSC model. This connection brings a fresh view to CNN, as we are able to attribute to this architecture theoretical claims such as uniqueness of the representations throughout the network, and their stable estimation, all guaranteed under simple local sparsity conditions. Lastly, identifying the weaknesses in the above scheme, we propose an alternative to the forward-pass algorithm, which is both tightly connected to deconvolutional and recurrent neural networks, and has better theoretical guarantees.

This talk was given at the workshop on Frame Theory and Sparse Representation for Complex Data, at the Institute for Mathematical Sciences (IMS) - National University of Singapore. This talk summarizes portions of the PhD work by my three PhD students, Vardan Papyan, Yaniv Romano, and Jeremias Sulam.
Style Transfer via Texture Synthesis
March 19th, 2017
Hebrew University

Style-transfer is a process of migrating a style from a given image to the content of another, synthesizing a new image which is an artistic mixture of the two. Recent work on this problem adopting Convolutional Neural-networks (CNN) ignited a renewed interest in this field, due to the very impressive results obtained. There exists an alternative path towards handling the style-transfer task, via generalization of texture-synthesis algorithms. I will present a novel such style-transfer algorithm that extends the texture-synthesis work of Kwatra et. al. (2005), while aiming to get stylized images that get closer in quality to the CNN ones.

This talk was given in the computer vision seminar in the Hebrew University. This is a joint work with Peyman Milanfar - Google Research.