ICRI-CI Israel Retreat 2016 – Presentations

Day 1, Tuesday, 24-May-2016

Opening Notes Ronny Ronen
Deep Learning 
Deep Learning – Capstone Overview  Shai Fine
Graying the Black Box of Deep Q Network Shie Mannor
Irreducible Representations of Symmetries and Solvable Deep Learning Models Naftali Tishby
Minimizing the Maximal Loss: Why and How  Shai Shalev Shwartz
On the Quest for a Deeper Understanding of Deep Learning Amnon Shashua
Geometric Structure of Multilayer Neural Networks Ohad Shamir
Geometry, Optimization and Regularization in Deep Learning Nati Srebro
RNN Fisher Vectors for Action Recognition and Image Annotation Lior Wolf
Arch/ML Brainstorm Report Shai Fine
Architecture 
Architecture – Capstone Overview  Debbie Marr
To Process-in-Storage, or not to Process-in-Storage?
That is the question
Uri Weiser
From Processing-in-Memory to Processing-in-Storage Ran Ginosar
Similarity Calculations in Resistive Memories Yuval Cassuto
On the Limitations of using Neural Networks for Memory Prefetching Yoav Etsion

Day 2, Wednesday, 25-May-2016                

Conversational Understanding  
Conversational Understanding – Capstone Overview Moshe Wasserblat
Integrated Pattern-Based and Distributional Learning of Lexical Relations with Recurrent Neural Networks Ido Dagan
Improving Speech Recognition with Discriminative Fluency Classification Moshe Koppel
Semisupervised learning with variational probabilities Amir Globerson
Visual
Session Overview Ronny Ronen
Strategical Argumentative Agent for Human Persuasion Sarit Kraus
PatchBatch: a Batch Augmented Loss for Optical Flow Lior Wolf
Statistics of RGBD Images Yair Weiss
Calibration of Multiple Cameras using Scene Dynamics
<PDF only, contact PI for full presentation >
Shmuel Peleg