Zoom Link: https://github.com/giorgianb/spdhackspring2021/blob/main/bit.ly/spdhack…
ΣΦΔ Hack Spring 2021 is ΣΦΔ's first annual machine learning hackathon. ΣΦΔ Hack Spring 2021 aims to introduce Stony Brook students to the rich and challenging field of machine learning, and develop the skills necessary to build sophisticated machine learning models on their own.
More info here: https://github.com/giorgianb/spdhackspring2021/blob/main/README.md
CG Group member (and SBU faculty) Chao Chen will speak on Fri, March 12, about the use of topological data analysis in machine learning for image analysis.
Chao has shared some of his research with the CG Group previously, and this will be a great opportunity to learn more about this exciting research area related to computational geometry/
Time: Friday, March 12, 2pm-3pm
Place: Zoom
https://stonybrook.zoom.us/my/
Title: Learning with Topological Information - Image Analysis and Label Noise
Speaker: Prof. Chao Chen (SBU)
Abstract: Modern machine learning faces new challenges. We are
analyzing highly complex data with unknown noise. Topology provides
novel structural information to model such data and noise. In this
talk, we discuss two directions in which we are using topological
information in the learning context. In image analysis, we propose a
topological loss to segment and to generate images with not only
per-pixel accuracy, but also topological accuracy. This is necessary
in analysis of images of fine-scale biomedical structures such as
neurons, vessels, etc. Extracting these structures with correct
topology is essential for the success of downstream
analysis. Meanwhile, we discuss how to use topological information to
train classifiers robust to label noise. This is important in practice
especially when we are using deep neural networks which tend to
overfit noise. These results have been published in NeurIPS, ECCV,
ICML and ICLR.
Abstract
Driving intelligence test is critical to the development
and deployment of autonomous vehicles. The
prevailing approach tests autonomous vehicles in life-
like simulations of the naturalistic driving environment.
However, due to the high dimensionality of the
environment and the rareness of safety-critical events,
hundreds of millions of miles would be required to
demonstrate the safety performance of autonomous
vehicles, which is severely inefficient. We discover that
sparse but adversarial adjustments to the naturalistic driving environment, resulting in the
naturalistic and adversarial driving environment, can significantly reduce the required test
miles without loss of evaluation unbiasedness. By training the background vehicles to
learn when to execute what adversarial maneuver, the proposed environment becomes
an intelligent environment for driving intelligence testing. We demonstrate the
effectiveness of the proposed environment in a highway-driving simulation. Comparing
with the naturalistic driving environment, the proposed environment can accelerate the
evaluation process by multiple orders of magnitude.
ZOOM LINK: Meeting ID: 950 6760 3617; Passcode: 426506
https://stonybrook.zoom.us/j/95067603617?pwd=dXQybEprSkNlTFY3WHlWYjViUG…
Bio
Professor Henry Liu is a professor in the Department of Civil and Environmental
Engineering at the University of Michigan, Ann Arbor. He is also a Research Professor at
the University of Michigan Transportation Research Institute and the Director for the
Center for Connected and Automated Transportation (USDOT Region 5 University
Transportation Center). Prof. Liu conducts interdisciplinary research at the interface
between civil and mechanical engineering. Specifically, his scholarly interests concern
traffic flow monitoring, modeling, and control, as well as testing and evaluation of
connected and automated vehicles. He has published more than 100 refereed journal
papers and is listed as one of the top 50 leading authors in the past 50 years (1969-2019)
in the prestigious Transportation Research journal. Professor Liu and his work have been
widely recognized in public media for promoting smart transportation innovations. He has
appeared on media outlets including CNBC, Forbes, Technode, etc. In 2019, Professor
Liu was invited to testify on national transportation research agenda in front of the US
House Subcommittee on Research and Technology. Professor Liu has nurtured a new
generation of scholars, and some of his PhD students and postdocs have joined first class
universities such as Columbia University, Purdue University, RPI, etc. Prof. Liu is the
managing editor of Journal of Intelligent Transportation Systems.
Date of Event
Joel H. Saltz, MD, PhD
SUNY Distinguished Professor Cherith Professor and Founding Chair
Department of Biomedical Informatics
Stony Brook University
Apostolos K. Tassiopoulos, MD, FACS
Professor of surgery and vice chair for quality and outcomes Chief of the Division of Vascular and Endovascular Surgery
Director of the Stony Brook Vascular Center Stony Brook Medicine
Title: Clinical applications of artificial intelligence to improve diagnosis and risk stratification for patients with aortic aneurysms
Time: Wednesday, Feb 17, 2021 3 pm - 4 pm
Join Zoom Meeting
https://stonybrook.zoom.us/j/95617197636?pwd=KytzZ2pVRG9SZGpKZUtpNXJISj...
Meeting ID: 956 1719 7636 Passcode: 924293
Title: Class visual similarity based noisy sample removal in generative Few Shot Learning
Time: Thursday, Feb 4, 11:30am - 1:00pm
Zoom:
https://stonybrook.zoom.us/j/
Meeting ID: 856 364 6526
Passcode: 203791
Abstract:
Over the past decade, larger datasets, hardware accelerations, and network architecture improvements have contributed to phenomenal achievements in many tasks of computer
vision. However, in the absence of large datasets, computer vision models struggle to learn
general representations which results in poor performance. Few-shot learning tries to address
this problem by proposing models which learn from a few examples.
I first give an overall review of few-shot learning methods. I particularly focus on generative Few Shot Learning(FSL) methods, which augment the scarce categories in a dataset by generating samples for those rare categories. As the actual class distribution can be complex and lie very close to each other, the sample generated for one class can be noisy or lie close to another class. However, none of the current FS generative methods perform any form of quality control of the generated samples.
In this work, I propose to identify and remove the generated samples that are less likely to be in the distribution of the few-shot class. Here I particularly deal with few-shot scenarios where the
prior information of the relationship between the classes based on visual similarity is available. The main idea is to exploit these priors to better identify the unreliable generated samples.
Particularly, I have proposed two methods based on class relationship to detect noisy generated samples. In the first method, we assume that the embedding space of each class follows a Gaussian distribution. From this assumption, I propose Gaussian Neighborhood (GN), a method to estimate how likely a generated sample is drawn from the estimated distribution of a few-shot class. We evaluate this method on the Hematopoiesis dataset. By simply eliminating samples based on thresholding our proposed GN scores, the few-shot classification performance is improved by 5% and 2% in five shot and one shot respectively, compared to the model trained on all generated images.
The GN scores represent the similarity distances from the generated samples to their classes, based on the assumption that each class is a Gaussian distribution. However, this assumption might be strict in many scenarios since the real distributions of data can be arbitrarily complex. Thus in my second proposed method, I aim to learn such similarity distances directly from data via metric learning. I propose to train a deep-network to regress the similarity distance between a pair of samples. This network is trained using both the class-level visual similarity information and the class labels. This method improves the 1-shot and 5-shot classification performances by 0.5% and 1% respectively, compared to GN.
Jerome Liang, PhD
Professor of Radiology, Biomedical Engineering, Electric and Computer Engineering, and Computer Science
Co-Director of Research
Department of Radiology
Artificial intelligence, machine learning and computer-aided diagnosis in cancer Imaging
February 11, 2021
12:00pm - 1:00pm
Virtual Seminar - Zoom
https://stonybrook.zoom.us/j/
Meeting ID: 981 5562 9970
Passcode: 950410
Host:
Wei Zhao, PhD
Professor of Radiology and Biomedical Engineering
Educational Objectives
Upon completion, participants should be able to:
(1) Learn different medical image representations of cancer attributes, such as heterogeneity, high tendency to grow, etc.
(2) Learn how computer (machine) can be trained (or programmed) to recognize the image representations.
(3) Learn how artificial intelligence can drive the machine learning to maximize the performance of computer-aided diagnosis (CADx).
Disclosure Statement
In compliance with the ACCME Standards for Commercial Support, everyone who is in a position to control the content of an educational activity provided by the School of Medicine is expected to disclose to the audience any relevant financial relationships with any commercial interest that relates to the content of his/her presentation.
The speaker, Jerome Liang, PhD, the planners; and the CME provider have no relevant financial relationship with a commercial interest (defined as any entity producing, marketing, re-selling, or distributing health care goods or services consumed by, or used on, patients), that relates to the content that will be discussed in the educational activity.
CONTINUING MEDICAL EDUCATION CREDITS
The School of Medicine, State University of New York at Stony Brook, is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
The School of Medicine, State University of New York at Stony Brook designates this live activity for a maximum of 1.0 AMA PRA Category 1 Credits™. Physicians should only claim credit commensurate with the extent of their participation in the activity.
Should you be logging in Zoom by using your tablet or mobile device, please be sure to add your Full Name and/or Email for CME credit.
Time: Jan 26, 2021 03:00 PM Eastern Time (US and Canada)
All are welcome!
Zoom Meeting:
https://stonybrook.zoom.us/j/
Meeting ID: 938 1855 2212
Passcode: 802722
Title: Data-Driven Document Unwarping
Abstract: Capturing document images is a common way to digitize and record physical documents due to the ubiquitousness of mobile cameras. To make text recognition easier, it is often desirable to digitally flatten a document image when the physical document sheet is folded or curved. However, unwarping a document from a single image in natural scenes is very challenging due to the complexity of document sheet deformation, document texture, and environmental conditions. Previous model-driven approaches struggle with inefficiency and limited generalizability. In this thesis, I investigate several data-driven approaches to tackle the document unwarping problem.
Data acquisition is the central challenge in data-driven methods. I first design an efficient data synthesis pipeline based on 2D image warping and train DocUNet, the pioneering data-driven document unwarping model, on the synthetic data. A benchmark dataset is also created to facilitate comprehensive evaluation and comparison. To improve the unwarping performance by training on more realistic data, I introduce the Doc3D dataset and DewarpNet. Supervised by 3D shape ground truth in Doc3D, DewarpNet is significantly better than DocUNet. DocUNet and DewarpNet depend on the synthetic data for the ground truth deformation annotation. To exploit the real-world images, I propose PaperEdge, a weakly supervised model trained with in-the-wild document images with easy-to-obtain boundary information. PaperEdge surpasses DewarpNet by utilizing both the synthetic data and weakly annotated real data in the Document In the Wild (DIW) dataset. Finally, I propose to incorporate the 3D physical constraints in training DewarpNet and PaperEdge. The constraints regulate the possible deformations on document papers. I also propose to augment the Doc3D and DIW dataset by introducing an online document segmentation model and better hardware.
A new virtual seminar series on Games, Decisions, and Networks will start this Friday. The series aims at bringing together researchers working on foundations and applications of games theory, decision theory, and networks from computer science, control, economics and operation research.
The advisory board for the series comprises Asu Ozdaglar (MIT), Christos Papadimitriou (Columbia), Drew Fudenberg (MIT), Eva Tardos (Cornell), Matthew O. Jackson (Stanford), Ramesh Johari (Stanford), and Tamer Başar (UIUC).
The first talk will be given by Costantinos Daskalakis (MIT) on January 22nd at noon ET, titled Equilibrium Computation and the Foundations of Deep Learning. Upcoming speakers include
- Rakesh Vohra (Upenn)
- Sanjeev Goyal (Cambridge)
- Aaron Roth (Upenn)
- Aislinn Bohren (Upenn)
- Jason Marden (UCSB)
and more to be added!