Program

Guy Tsafrir
VP R&D
Cognata

Bio

Guy Tsafrir is the VP R&D of Cognata and has a strong track record in designing complex multidisciplinary systems.
Before joining Cognata, he held the position of digital health team leader at the Samsung Strategy and Innovation Center in Israel.
During his time at GE Healthcare, he served in various positions including software team leader, control & algorithms team leader, software & system manager, and R&D manager. Before joining GE Healthcare, he developed software and algorithms at Versamed and worked as a software engineer at Verint. Guy holds a BSc in Physics from Tel Aviv University.

Abstract

The greatest challenge regarding simulation is finding the matrix that compares it to real life. In this session, Danny will present a mathematical matrix that defines this relation and show how a proper simulation can be constructed based on deep learning techniques. He will also supply live examples of the Cognata simulation platform.

Dr. Leonid Karlinsky
DL Team Lead, CVAR Group
IBM Research AI

Bio

Dr. Karlinsky leads the CV & DL research team in the Computer Vision and Augmented Reality (CVAR) group @ IBM Research AI. He is a Computer Vision and Machine Learning expert with years of hands on experience. He is actively publishing research papers in leading CV and ML venues such as ECCV, CVPR and NIPS and is actively reviewing for these conferences for the past 10 years. Dr. Karlinsky holds a PhD degree in CV from the Weizmann Institute of Science, supervised by Prof. Shimon Ullman.

Abstract

Learning to classify and localize instances of objects that belong to new categories, while training on just one or very few examples, is a long-standing challenge in modern computer vision. This problem is generally referred to as 'few-shot learning'. It is particularly challenging for modern deep-learning based methods, which tend to be notoriously hungry for training data.  In this talk I will cover several of our recent research papers offering advances on these problems using example synthesis (hallucination) and metric learning techniques and achieving state-of-the-art results on known and new few-shot benchmarks. In addition to covering the relatively well studied few-shot classification task, I will show how our approaches can address the yet under-studied few-shot localization and multi-label few-shot classification tasks. In addition to this talk, a detailed tutorial covering the few-shot learning field in general will be given by Dr. Joseph Shtok from my team.

Eran Avidan 
Senior Software Engineer
Intel

Bio

Eran Avidan is a senior software engineer in Intel’s Advanced Analytics Department. Eran enjoys everything distributed, from Spark and Kafka to Kubernetes and TensorFlow. He holds an MS in computer science from the Hebrew University of Jerusalem.

Abstract

Deep learning has recently become an abundant technology for analyzing video data. However, the increasing resolution and frame rates of videos make real-time analysis a remarkably challenging task.

We present an overview of a novel architecture based on Redis, Docker, and TensorFlow that enables real-time analysis of high-resolution streaming video. The solution can serve advanced deep learning algorithms at subsecond rates and appears fully synchronous to the user despite containing an asynchronous backend. The talk offers a demo using visual inspection and shares results that highlight the solution’s applicability to real-time neural network processing of videos. The approach is generalizable and can be applied to diverse domains that require video analytics.

Daniel Urieli
Senior Research Scientist
GM Research

Bio

Daniel Urieli is a senior research scientist at GM Research, applying Reinforcement Learning for challenging problems in the world of transportation. Daniel received his Ph.D. in artificial intelligence from The University of Texas at Austin, where he was advised by Prof. Peter Stone. During his Ph.D. studies, Daniel won several awards, including 1st place in the international RoboCup competition, 1st place in the international Power Trading Agent Competition, Best Contribution Award at NIPS workshop on Machine Learning for Sustainability, and an NSF IGERT Fellowship. Daniel received his M.Sc and his B.Sc in computer science and mathematics from Tel-Aviv University.

Abstract

Reinforcement learning (RL) is a branch of machine learning concerned with using experience gained through interacting with the world and evaluative feedback to improve a system's ability to make behavioral decisions [Littman, Nature 2015].  In recent years, RL achieved some major successes, which include: super-human GO playing, challenging robotic manipulation tasks execution, super-human computer game playing, and autonomous helicopter flight. This tutorial will start from RL's foundations, continue with some of RL's core techniques, and will cover some of RL's recent successes. This tutorial assumes no prior background in RL - general CS/math background will suffice.

Oron Nir
Senior Data Scientist
Microsoft Media AI Research Team

Bio

Oron is a founding member of the Microsoft Video Indexer, a media analytics cloud service recently announced GA, and a Senior Data Scientist at the Microsoft Media AI research team. He has acquired his B.Sc. and M.Sc. degrees at the Ben-Gurion University of the Negev, with a thesis at the field of Computer Vision.

Abstract

Media companies accumulate large archives and struggle to transform it into business value. Media is unstructured and therefore hard to manage at scale. Content categorization by topics is an intuitive approach that makes it easier for people to search. Many organizations turn to tagging their content manually, which is expensive and isn’t scalable. To automate this process, we apply multimodal topic detection and tracking (TDT). The information communicated through the speech, OCR, and identified celebrities is detected using Named Entity Recognition, clustered by concepts into similarity graphs, named using Wikipedia Categories ranker in an interpretable manner with evolvement over time.

Soliman Nasser
Lead Research and Development Engineer
Blink Technologies Inc.

Bio

Soliman Nasser is a Computer Scientist and Researcher, working as a Lead Research and Development Engineer at Blink Technologies Inc. Previously, Soliman was a part of the Algorithms team at Intel RealSense, developing depth cameras 3D core algorithms. Soliman completed his B.Sc in tandem with high school studies as a part of "Etgar" Program, and M.Sc in Computer Science under the supervision of Dr. Dan Feldman at the Robotics and Big Data Lab, both from the University of Haifa. Soliman's research interests include: Computer Vision, Multiple View Geometry, {Machine, Deep}-Learning, Vision for Robotics, SLAM, Depth cameras , Eye-Tracking and more.

Abstract

The eyes play a vital role in the perception of our evolving surrounding as well as in communication with other human beings. Apart from the behavioral and scientific research in medical diagnoses
and psychological studies, estimating gaze direction is also important in Mixed Reality, AI assistant devices, Automotive, Human-Computer and Robot interaction, User authentication and many other applications.
In this talk, I will give a short introduction to eye-tracking research, compare between the geometric (model based) and data-driven (appearance based) methods, with emphasis on the ad-vantages and disadvantages of deep-learning in this domain and present some interesting usages of this technology.

Dolev Pomeranz M.Sc.
Chief Architect and Head of Research
Trax

Bio

Dolev Pomeranz is Chief Architect and Head of Research at Trax, a startup in the retail industry aiming to digitize the physical world of retail. There he works on both algorithmic and engineering challenges. He holds his M.Sc. from Ben-Gurion University. His thesis was in the field of computational Jigsaw puzzle solving.

Abstract

Augmented Reality is the combination of Computer Vision and Computer Graphics. As technology mature and is now being commoditized, new innovation can emerge. In this talk, we would share our story of Augmented Reality @ Trax. How we use it for multi-perspective geometry, indoor mapping and navigation, and even challenging the shopping experience.

Asi Shefer
Lead Data Scientist 
LogMeIn

Bio

Asi Shefer is a lead data scientist at LogMeIn’s AI research center. His research focuses on natural language processing and computer vision with a spatial interest in multimodal frameworks where different modalities represent similar objects. Asi holds his M.Sc from Ben-Gurion University.

Abstract

Multi-domain problems like grounding text in the Visual world are crucial for many real world applications, yet remain mostly unsolved.

We examine the effects of combining visual and linguistic representations revealing their fundamentally different nature yielding an imbalanced co-adaptation.

We introduce Cross Domain Normalization to dramatically stabilize learning, reduce overfit and speed up learning (up to 19X faster than Batch Normalization) though manipulation of cross domains statistics. With CDN our extremely simple model significantly outperforms all today’s SOTA models. The insights gained by investigating the linguistic and visual co-adapted parameters can be utilized for other multi-domain tasks and co-adaptation in general.

Yonatan Cohen, Ph.D.
CTO
Quantum Machines

Bio

I completed my Ph.D. in Physics at the Weizmann Institute in professor Moty Heiblum's group, where I investigated quantum electronics devices and topological quantum states. 

A year ago, together with Itamar Sivan and Nissim Ofek, I co-founded Quantum Machines (QM), which is the first Israeli startup in quantum computing. At QM we develop control and operation systems for quantum computers. With a multidisciplinary team of highly motivated physicists and engineers, all passionate about solving the challenges of quantum control, we hope to help making the dream of a large-scale quantum computer become a reality. 

Abstract

Quantum mechanics, which was born at the beginning of the 20th century and revolutionized our understanding of nature, implies that nature is far more counter-intuitive and far richer than anyone could imagine. In the 1980s, physicists understood that this richness of nature can be used to construct new computers that can outperform any classical computer. In the decades that followed, physicists have made huge progress in designing and controlling quantum systems, and today quantum computers seem to be closer than ever. What are the basic principles of quantum computers? how do they allow them to beat classical computers? What problems will they be best for? 

Ofir Bibi
Director of Research
Lightricks

Bio

Ofir Bibi is the director of research at Lightricks. His research is in the fields of machine learning, statistical signal processing, computational photography and computer graphics. He has taken leading research positions in building systems for estimation and prediction of various signals from weather, electricity consumption and financial markets to photos, videos and auditory signals. Currently he is focused on ways to create consistent results from neural networks in a resource constraint environment.

Abstract

Who hasn't gotten an image or video with some dull skies taking up most of the space?

Professionals may edit these videos using advanced and time-consuming tools, to replace the sky with a more expressive or imaginative sky. 

In this work, we propose an algorithm for automatic replacement of the sky region in a video with a different sky.

The method is fast, achieving close to real-time performance on mobile devices and can be fully automatic.

Joint work with Tavi Halperin, Harel Cain and Michael Werman. To be presented at Eurographics 2019.

Liat Zakay
CEO
Donde Search

Bio

Donde Search CEO & Founder, Donde award-winning AI Computer Vision and NLP technology mimics the way people think about products helping retailers read their customers' mind. Liat holds a BSc in a Computer Science and Economics student from BGU of the Negev. Liat, is a tech entrepreneur. She led Magshimim, a Cyber Security Training program in Israel peripheral area that became a national success and was head of Cyber & Computer Networks course. Prior to her academic studies, Liat was a team leader and served as an officer in an elite technological unit in the Israeli Intelligence and managed projects using cutting edge technologies and high significance and impact.

Roni Gurvich
Head of Algorithms
Donde Search

Bio

Roni is leading the Algo team at Donde Search, developing machine learning technology that uses NLP and CV to help retailers better understand their products and customers, and enable customers seamless navigation through massive catalogs.

Roni holds a B.Sc in biomedical engineering from Tel Aviv University and for the last 6 years he has been creating AI products using state of the are models in the E-Commerce, automotive and medical fields.

Abstract

Donde Search uses Computer Vision and Natural Language Processing to analyze product pages and extract meaningful information to improve navigation, merchandising, personalization, and search across e-commerce platforms.

In this talk, we will present some of Donde's solutions and discuss their underlying technologies. We will show how Donde's technology transforms a challenging and sometimes frustrating search experience into a natural and engaging visual process that allows humans to navigate their way in a database including hundreds of thousands of items. We will discuss the gap between state-of-the-art deep-learning algorithms and real working solutions for enterprises, including some of the most common problems in transforming working algorithms into product like handling structured data, combining texts and images and using unsupervised methods for data-augmentation and for allowing better training and learning.

Eldad Klaiman
Principal Scientist
Roche Innovation Center Munich

Bio

B.Sc and M.Sc in electrical engineering from the Technion IIT in Haifa. For the past 4 years working as a principal scientist at the Roche Innovation Center near Munich, Germany in the Oncology division of Roche Pharma Research and Early Development. During this period he has been responsible for the development and application of machine learning algorithms, mostly in the field of pathology tissue analytics and leveraging Roche's internal datasets in order to develop novel technology and techniques for informing and supporting oncology drug development.

Abstract

Modern-day pharma organizations face mounting challenges in the field of drug research and development due to the pharma competitive landscape, the complexity of treatment methods and increasing number of drug combination studies. At Roche we are pursuing a Personalized Healthcare (PHC) approach to drug development, which is meant to match the best medicine to each specific patient. This approach has the potential to drastically improve drug efficacy for our patients but it also poses the challenge of patient-drug selection. The amount of data available in the drug development process, from histology and clinical imaging, from genomics and sequencing, from real-world data and from diagnostic methods present a unique opportunity for the extraction of novel insights and the development of new technologies. These novel machine learning techniques are helping us address these challenges and leverage this information to assist scientists in the research process, supporting decision making and bringing better drugs to patients, faster. 

Samah Khawaled
Technion – Israel Institute of Technology

Bio

Samah Khawaled graduated from Technion’s Viterbi faculty of electrical engineering in 2017. She is currently a graduate student working on her research towards M.Sc degree. Her research is concerned with modeling of Natural Stochastic Textures (NST) that incorporate also structural information and with the application of the model in analysis and classification of images. Samah supervises image-processing-related projects and she is involved as a T.A in various courses. Previously she worked at Intel and served as a tutor in Landa (equal opportunities) project. Samah is the recipient of the Israeli Ministry of Science and Technology Fellowship for 2017-2019.

Abstract

Natural Stochastic Textures (NST) images exhibit self-similarity and Gaussianity that are the two main properties characteristic of Fractional Brownian Motion (fBm) processes. We consider non-pure NST images that contain also structural information. The latter is characterized by having profound local phase information, whereas the former is characterized by random spatial phase. In this meeting, we address primarily the fractal-based layer of the model and implementing it on the NST component. We also discuss applications where the approach can be used, with special emphasis on medical images. Examples of mammography and bone X-ray images are presented.

* Join work with Prof. Yehoshua Y.Zeevi

Ortal Senouf, M.Sc
Researcher at VISTA lab
Technion & MaxQ-ai

Bio

Ms. Ortal Senouf is a recent MSc graduate from the Department of Electrical Engineering of the Technion – Israel Institute of Technology. During her master’s studies she has been dividing her time between her research at the VISTA lab of the Department of Computer science at the Technion, and the algorithm research team of MaxQ-AI, a medical AI start-up she had joined as first employee, soon after completing her B.Sc. in 2013 from the Department of Bio-Medical Engineering. Among Ms. Senouf research interests: Medical imaging- analysis and acquisition, machine learning and computer vision. 

Abstract

Viewing ultrasound imaging (US), or any other medical imaging modality for that matter, as an inverse problem, in which a latent image is reconstructed from a set of measurements, current research is focused mostly on learning the inverse operator producing an image from the measurements. The scope of our work differs sharply in the sense that we propose to learn the parameters of the forward model, specifically, the transmitted beams patterns (Tx), together with the receive beamforming (Rx). We demonstrate a significant improvement in the image quality compared to the standard patterns used in standard fast US acquisition settings.

Rana Hanocka
Ph.D. Student
Tel Aviv University

Bio

Rana Hanocka is a Ph.D. student under the supervision of Daniel Cohen-Or and Raja Giryes at Tel Aviv University. Her research is focused on applying convolutional neural networks on irregular data. 

Abstract

Polygonal meshes provide an efficient representation for 3D shapes. They explicitly capture both shape surface and topology, and leverage non-uniformity to represent large flat regions as well as sharp, intricate features. This non-uniformity and irregularity, however, inhibits mesh analysis efforts using neural networks that combine convolution and pooling operations. In this paper, we utilize the unique properties of the mesh for a direct analysis of 3D shapes using MeshCNN, a convolutional neural network designed specifically for triangular meshes. Analogous to classic CNNs, MeshCNN combines specialized convolution and pooling layers that operate on the mesh edges, by leveraging their intrinsic geodesic connections. Convolutions are applied on edges and the four edges of their incident triangles, and pooling is applied via an edge collapse operation that retains surface topology, thereby, generating new mesh connectivity for the subsequent convolutions. MeshCNN learns which edges to collapse, thus forming a task-driven process where the network exposes and expands the important features while discarding the redundant ones. We demonstrate the effectiveness of our task-driven pooling on various learning tasks applied to 3D meshes.

Tzofnat Greenberg-Toledo
Ph.D. student
Technion – Israel Institute of Technology

Bio

Tzofnat Greenberg-Toledo is a Ph.D. student at the Andrew and Erna Viterbi Faculty of Electrical Engineering, Technion – Israel Institute of Technology. She received her B.Sc. degree from the Andrew and Erna Viterbi Faculty of Electrical Engineering at the Technion in 2015. From 2014 to 2016 she worked as a logic design engineer at Intel corp. As of 2016, she is a graduate student in Electrical Engineering at the Technion. Her current research is focused on computer architecture and accelerators for Deep Neural Networks with the use of memristors.

Abstract

Deep Neural Networks (DNNs) are usually executed by commodity hardware, mostly FPGA and GPU platforms, and accelerators, such as Google's TPU. However, when executing DNN algorithms, the conventional von Neumann architectures, where the memory and computation are separated, pose significant limitations on performance and energy efficiency, as DNN algorithms are compute and memory intensive.  Emerging memory technologies, known as memristors, enable in-place, highly parallel, and energy efficient analog multiply-accumulate operations. This is known as processing-near-memory (PNM). This talk will present the potential and opportunities of integrating memristors into DNN accelerators design.

Dr. Joseph Shtok
Research Staff Member at Computer Vision and Augmented Reality Team
IBM Research AI, Haifa Lab

Bio

Joseph Shtok holds an M.Sc. degree in Mathematics and a Ph.D. degree in Computer Science, from the Technion – Israel Institute of Technology. His areas of expertise include Machine Learning, Computer Vision with focus on Object Recognition, Image Processing, Medical Imaging and depth sensors. Joseph is one of the leading research scientists of the IBM Research AI team, specializing in object recognition and 3D algorithms. His research papers are published in leading Computer Vision and Signal Processing conferences such as CVPR, NIPS, ISBI, ICASSP.

Abstract

Few-shot learning is a rapidly evolving field in computer vision, where the standard tasks of classification and detection are addressed in a challenging scenario with only few examples per class available for training. Using the approaches of meta-learning, metric learning, synthesis (hallucination) and data augmentation, substantial progress has been made in this field during the recent years. We will provide an overview of the state of the art in few-shot learning, highlighting and describing the main methods and results in the aforementioned directions.

Ayelet Akselrod-Ballin, PhD
Director of AI
Zebra Medical Vision

Bio

Ayelet Akselrod-Ballin’s is the director of AI at  Zebra Medical Vision where she leads an excellent team of CV-ML researchers and data scientists. Her work focuses on developing novel technologies for computer vision, machine learning, deep learning, and biomedical image analysis. Ayelet did her Post-doctoral research as a fellow in the Computational Radiology Laboratory at Harvard Medical School, Children’s Hospital (Boston) and she holds a Ph.D. in Applied Mathematics and Computer Science from Weizmann Institute of Science. Prior to joining Zebra Medical Vision,  Ayelet lead the medical imaging research technology at IBM-research and lead the Computer Vision & Algorithms team at MOD.

Abstract

The advances in deep learning algorithms together with the continuous increase in digital information volume provide a unique opportunity in the healthcare domain. Recent deep Learning studies have established remarkable performance in complex diagnostic tasks.  Nevertheless, some of the challenges include the need for interpretability to ease the human-machine collaboration, together with integration of imaging data and other types of data such as text, which is complex, noisy, heterogeneous and sometimes missing. In this talk we discuss several key applications spanning neuro CT, Chest X-ray and Spinal analysis with CT and demonstrate the results scaling large multi-domain dataset.

Jacob Gildenblat
Co-Founder and CTO
DeePathology.ai

Bio

DeePathology.ai develops digital pathology products for diagnostics and for pharma research.

Before DeePathology, Jacob was the leader of the deep learning group at SagivTech.

Jacob has a BSc and MSc in Electrical Engineering from Tel Aviv university. 

Abstract:

Common problems in the process of developing AI solutions for the medical field are highly unbalanced datasets on one hand and limited annotation resources on the other hand.

The use of Active Learning can dramatically help with both issues.

The task of Cell Detection is very important in digital pathology. For example, analyzing the quantity and density of immune cells can provide important indications on the progress of cancer.

This is a tedious task when manually done by pathologists and thus, automating this process is desirable.

Automating cell detection requires annotating large amounts of data, which is usually very unbalanced.

The DeePathology.ai Cell Detection Studio is a do it yourself tool for pathologists to train deep learning cell detection algorithms on their own data.

Using this tool, deep learning cell detection solutions can be easily created by the pathologist very quickly.

In the talk we will use the example of the DeePathology.ai Cell Detection Studio to demonstrate how Active Learning can be used for medical imaging annotation.

We will also present our approach for using active learning with unbalanced datasets.

Dr. Refael Vivanti
Researcher​
Rafael

Bio

Dr. Refael Vivanti is a researcher at the Computer Vision for Autonomous Systems department at Rafael Advanced Defense Systems Ltd. Dr. Vivanti holds a Ph.D. degree in medical computer vision from the Hebrew University in Jerusalem, Israel, supervised by Prof. Leo Joskowicz. His research areas are deep reinforcement learning, deep learning for autonomous systems and SLAM.

Abstract

We succeeded to train an agent to drive in a specific unvisited area using only a realistic 3D model of the area which was built using merely aerial images. The training system is end to end: the inputs are realistically rendered images from the 3D model to the agent perspective and the outputs are driving commands to the moving platform. While training, the system got only sparse weak supervision from collision detections with the 3D model. Our main novelty is the separate treatment to the two linked challenges: navigation and obstacle avoidance, which both affect the platform location and position, but have different rewards distribution and thus hard to learn simultaneously. The two are approached using different methods and combined in the final agent behavior. We hope to use the trained agent for driving a real vehicle inside the modeled area. 

Adham Ghazali
CEO & Co-Founder
Imagry

Bio

Adham Ghazali is the CEO of Imagry. He spent the last 10 years working on various machine learning problems including large scale computer vision, Brain-computer interfacing, and Bio-Inspired facial recognition. He is interested in the intersection between biology and computer science. At his current post, he is responsible for strategic R&D and Business Development.  
Prior to cofounding Imagry, Adham was a brain researcher focusing on the study of the visual system in infants. 

Abstract

Autonomous driving has come a long way albeit it's success is limited only to those areas where a High-Definition map is pre-built and known. Inspired by the recent achievements in Artificial Intelligence particularly of methods that combine trees and DNNs (e.g. Alpha Go Zero), this talk demonstrates how to effectively combine Deep Learning and convention path planning to drive in unknown areas. 

Dr. Moti Freiman
Staff Research Scientist
Global Advanced Technology, CT/AMI
Philips Healthcare

Bio

Moti Freiman is a staff research scientist at Philips Healthcare where he is developing advanced algorithms with the aim of improving the capacity of medical imaging devices to provide clinically meaningful information by leveraging machine learning, computer vision and image processing algorithms.

Prior to Philips, Dr. Freiman was an Instructor in Radiology at Harvard Medical School where he developed advanced algorithms for quantitative analysis of diffusion-weighted MRI data.

Dr. Freiman is the recipient of an NIH R01 research grant and the 2012 Crohn's and Colitis foundation of America research fellow award. He is the author and co-author of more than 40 journal and full-length conference papers and holds several patents and patent applications.

Abstract

Deep sparse auto-encoders with mixed structure regularization (MSR) in addition to explicit sparsity regularization term and stochastic corruption of the input data with Gaussian noise have the potential to improve unsupervised abnormality detection. Unsupervised abnormality detection based on identifying outliers using deep sparse auto-encoders is a very appealing approach for medical computer aided detection systems as it requires only healthy data for training rather than expert annotated abnormality. 

In the task of detecting coronary artery disease from Coronary Computed Tomography Angiography (CCTA), our results suggests that the MSR has the potential to improve overall performance by 20-30% compared to deep sparse and denoising auto-encoders.

Shmoolik Mangan
Algorithms Development Manager
VayaVision

Bio

Shmoolik Mangan, is leading the algorithms development of VayaVision, a startup focused on algorithms solutions for autonomous driving.

Has Ph.D. from Weizmann inst., and 27 years of experience and leadership in the fields of Algorithms, Physics, Optics and Systems development of multi-disciplinary systems. Developed products for Metrology, Inspection, Detection and Classification systems for semiconductors and electronics manufacturing industries. Worked in Applied Materials and Orbotech.

Abstract

The foundations of autonomous driving (AD) are based on advanced sensors and algorithms for environmental perception, especially for road and obstacles detection. The sensor set for AD is expected to contain a combination of low and high resolution, image and distance sensors, including cameras, Radars and Lidars. Low-level sensor fusion uses all sensor to generate a high-density pixel-level joint image-distance HD-RGBd model through upsampling. The environmental perception is generated by processing this joint HD-RGBd model through unified algorithms. While seems prone to loss of one sensor in the set, we actually developed built-in redundancy mechanisms that compensate for the loss of one sensor through alternative low-level fusion and detection of all the remaining sensors, with known loss of accuracy that can be used by the driving system to continue driving through at lower speed.

Assaf Shocher
Ph.D. Candidate
Weizmann Institute of Science

Bio

Assaf Shocher is a Ph.D. candidate at the Weizmann Institute of Science, in the Department of Computer Science and Applied Mathematics, under the supervision of Prof. Michal Irani. His research focuses on Deep-Learning and Computer Vision, especially on Unsupervised Methods for Neural Networks. 
He has received his MSc from the Weizmann Institute, with dean's prize for outstanding students and B.Sc in Physics and in Electrical Engineering from Ben-Gurion University. Assaf has co-founded a fintech startup and worked as a Machine Learning team leader and as a data scientist in several startups. Currently, Assaf is also a lecturer at the Deep Learning Academy TLV, teaching Machine Learning and Deep Learning courses. 

Abstract

Deep Learning has always been divided into two phases: Training and Inference. Deep networks are mostly used with large data-sets both under supervised (Classification, Regression etc.) or unsupervised (Autoencoders, GANs) regimes. Such networks are only applicable to the type of data they were trained for and do not exploit the internal statistics of a single datum.
We introduce Deep Internal Learning; We train a signal-specific network, we do it at test-time and on the test-input only, in an unsupervised manner (no label or ground-truth). In this regime, training is a part of the inference, no additional data or prior training is taking place. This is possible due to the fact that one single instance (be it image, video or audio) actually contains a lot of data when internal statistics are exploited. In a series of papers from the last year, that will be reviewed throughout the talk, I will demonstrate how we applied this framework for various challenges: Super-Resolution, Segmentation, Dehazing, Transparency-Separation, Watermark removal. I will also show how this approach can be incorporated into Generative Adversarial Networks by training a GAN on a single image for the challenge of retargeting.

Sagie Benaim
PhD Candidate
Tel Aviv University

Bio

A PhD candidate under the supervision of Prof. Lior Wolf at Tel Aviv University. Sagie has a several publications on GANs and related models, and its his main domain expertise.

 Abstract

The talk will focus and the recent advances in Generative Adversarial Networks (GANs) for Image to Image Translations. I will begin by describing the problem and the distinction between supervised and unsupervised approaches, focusing on the approaches of pix2pix, DTN, CycleGAN, Unit and DistanceGAN. I will then describe recent approaches which allow for variation in the target domain as well as "one-shot"-translation. Lastly, I will talk about the application of these techniques in other domains such as audio and text. 

Karina Odinaev
Founder, COO, Chief IP Officer
Cortica

Bio

Karina brings with her an extensive background in computational neuroscience and brain research. Karina pursued her award-winning research in the area of non-linear dynamical computational systems at the Technion, Israeli Institute of Technology, under the direction of Prof. Josh Zeevi.

Karina served as COO and VP Product of Cortica for 8 years. Karina has a vast experience in product characterization & development for high-dimensional and high-volume data processing systems. Prior to founding CORTICA, Karina was COO and VP Product of LCB Ltd., developing high-end systems for real-time voice recognition.

As Chief IP Officer, Karina’s accomplishments have helped Cortica generate an IP portfolio of over 200 patents and inventions and maintain its position as a top patent holder in Artificial Intelligence.

Karina served in an elite intelligence unit in IDF, leading a team of data production and analysis. Karina is part of a forum dealing with Israeli strategic homeland security issues.

Karina featured in the “Forty under 40” list of Israel in 2017, and one of the “top 20 leading entrepreneurs” in 2018. 

Abstract

The road to fully autonomous vehicles has been filled with challenges. When will cars drive themselves? What will it take from a technological standpoint to get us there? During this presentation, we will discuss the industry and technological challenges surrounding self-driving vehicles. We will illuminate how these hurdles can be overcome by moving from a deep learning model into a completely new paradigm of machine learning technology – Autonomous AI

Yuval Nirkin
Ph.D. student
Bar-Ilan University

Bio

Yuval is co-founder and CTO of DeepNen Ltd and a Ph.D. student in Bar-Ilan University under the supervision of Prof. Yosi Keller and Prof. Tal Hassner. Yuval has a BSc in Computer Engineering from the Technion and an MSc in Computer Science from the Open University of Israel, and he was an academic officer in one of the biggest technology units in the IDF. Yuval’s research is focused on Deep Learning based face synthesis.

Abstract

Face swapping is one of the most important face synthesis problems. It has numerous practical application such as hairstyle replacement, face spoofing, and data augmentation for machine learning. One of the most prominent face swapping methods is DeepFakes, which has recently attracted a lot of attention from the media around the world by making face swapping more accessible to regular users. The rapid progress in face synthesis should raise concerns on its implications on society. What will happen when those methods become easily accessible and harder to distinguish from real images?

Stav Shapiro
EE Ms.c Student
Technion

Bio

Stav Shapiro is a CTO at an AI research branch in the Defense Community. He double majored in Physics and Electrical Engineering in the Technion, and is currently completing his thesis under the supervision of Prof. Michael Elad.  His hobbies include playing video games, reading sci-fi novels and eating his wife's, Maya, amazing food.

Abstract

Patch matching is a key ingredient in many image processing and computer vision  applications. Learning base approaches for patch-matching were shown to be successful, but were tailored towards specific tasks. A more general approach for improving the PM engine was recently introduced by Romano et. al, showing the potential improvement in using their modified similarity measure on several PM-based tasks. In our research, we investigate a deep-learning based strategy that preserve the order between distances and show that the order-preserving approach can improve upon previous work by a significant margin.