PAII Inc., Bethesda Research Lab, USA
Bio:Le Lu received a PhD in 2007 from Johns Hopkins University. During his first six years at Siemens, he made significant contributions to the company’s CT colonography and Lung CAD product lines. From 2013 to 2017, Dr. Lu served as a staff scientist in the Radiology and Imaging Sciences department of the National Institutes of Health Clinical Center. He then went on to found Nvidia’s medical image analysis group and he held the position of senior research manager until June 2018. Since then, he has been the Executive Director at PAII Inc., Bethesda Research lab, Maryland, USA which has become one of the leading industrial research labs in medical imaging. He was the main technical leader for two of the most-impactful public radiology image dataset releases (NIH ChestXray14, NIH DeepLesion 2018). He won NIH Clinical Center Director Award in 2017, NIH Mentor of the year award in 2015, and won numerous best paper awards in MICCAI and RSNA.
Abstract: This talk describes a roadmap on three key problems in pancreatic cancer imaging solution: early screening, precision differential diagnosis, and deep prognosis on patient survival prediction. (1) Based on a new self-learning framework, we train the pancreatic ductal adenocarcinoma (PDAC) segmentation model using a larger quantity of patients (≈1,000, four institutions), with a mix of annotated/unannotated venous or multi-phase CT images. Pseudo annotations are generated by combining two teacher models with different PDAC segmentation specialties on unannotated images, and can be further refined by a teaching assistant model that identifies associated vessels around the pancreas. Our approach makes it technically feasible for robust large-scale PDAC screening from multi-institutional multi-phase partially-annotated CT scans. (2) We propose a holistic segmentation-mesh classification network (SMCN) to provide patient-level diagnosis, by fully utilizing the geometry and location information. SMCN learns the pancreas and mass segmentation task and builds an anatomical correspondence-aware organ mesh model by progressively deforming a pancreas prototype on the raw segmentation mask. Our results are comparable to a multimodality clinical test that combines clinical, imaging, and molecular testing for clinical management of patients with cysts. (3) Accurate preoperative prognosis of resectable PDACs for personalized treatment is highly desired in clinical practice. We present a novel deep neural network for the survival prediction of resectable PDAC patients, 3D Contrast-Enhanced Convolutional Long Short-Term Memory network (CE-ConvLSTM), to derive the tumor attenuation signatures from CE-CT imaging studies. Our framework can significantly improve the prediction performances upon existing state-of-the-art survival analysis methods. This deep tumor signature has evidently added values (as a predictive biomarker) to be combined with the existing clinical staging system.
Cornell University, USA
Bio: Kilian Weinberger is an Associate Professor in the Department of Computer Science at Cornell University. He received his Ph.D. from the University of Pennsylvania in Machine Learning and his undergraduate degree in Mathematics and Computing from the University of Oxford. During his career he has won several best paper awards, at ICML (2004), CVPR (2004, 2017), AISTATS (2005) and KDD (2014, runner-up award). In 2011 he was awarded the Outstanding AAAI Senior Program Chair Award and in 2012 he received an NSF CAREER award. He was elected co-Program Chair for ICML 2016 and for AAAI 2018. Currently he serves as an elected board member for the international machine learning society. In 2016 he was the recipient of the Daniel M Lazar '29 Excellence in Teaching Award. Kilian Weinberger's research focuses on Machine Learning and its applications, in particular on learning under resource constraints, metric learning, Gaussian Processes, computer vision and deep learning. Before joining Cornell University, he was an Associate Professor at Washington University in St. Louis and before that he worked as a research scientist at Yahoo! Research in Santa Clara.
Abstract: 3D object detection is an essential task in autonomous driving. Recent techniques excel with highly accurate detection rates, provided the 3D input data is obtained from precise but expensive LiDAR technology. Approaches based on cheaper monocular or stereo imagery data have, until recently, resulted in drastically lower accuracies - a gap that was commonly attributed to poor image-based depth estimation. In this talk we argue that it is not the precision or quality of the data, but its representation that accounts for the majority of the difference. Taking the inner workings of convolutional neural networks into consideration, we propose to convert image-based depth maps to pseudo-LiDAR representations - essentially mimicking the LiDAR signal. With this representation we can apply different existing LiDAR-based detection algorithms. Our approaches yield impressive improvements on the popular KITTI 3D object detection benchmark data set, and have already been adapted by the self-driving car community.
Swiss Federal Institute of Technology in Lausanne, Switzerland
Bio: Volkan Cevher received the B.Sc. (valedictorian) in electrical engineering from Bilkent University in Ankara, Turkey, in 1999 and the Ph.D. in electrical and computer engineering from the Georgia Institute of Technology in Atlanta, GA in 2005. He was a Research Scientist with the University of Maryland, College Park from 2006-2007 and also with Rice University in Houston, TX, from 2008-2009. Currently, he is an Associate Professor at the Swiss Federal Institute of Technology Lausanne and a Faculty Fellow in the Electrical and Computer Engineering Department at Rice University. His research interests include machine learning, signal processing theory, optimization, and information theory. Dr. Cevher is an ELLIS fellow and was the recipient of the Google Faculty Research Award on Machine Learning in 2018, IEEE Signal Processing Society Best Paper Award in 2016, a Best Paper Award at CAMSAP in 2015, a Best Paper Award at SPARS in 2009, and an ERC CG in 2016 as well as an ERC StG in 2011.
Abstract: Thanks to neural networks (NNs), faster computation, and massive datasets, machine learning (ML) is under increasing pressure to provide automated solutions to even harder real-world tasks beyond human performance with ever faster response times due to potentially huge technological and societal benefits. Unsurprisingly, the NN learning formulations present a fundamental challenge to the back-end learning algorithms despite their scalability, in particular due to the existence traps in the non-convex optimization landscape, such as saddle points, that can prevent algorithms to obtain “good” solutions.
In this talk, we describe our recent research that has demonstrated that the non-convex optimization dogma is false by showing that scalable stochastic optimization algorithms can avoid traps and rapidly obtain locally optimal solutions. Coupled with the progress in representation learning, such as over-parameterized neural networks, such local solutions can be globally optimal.
Unfortunately, this talk will also demonstrate that the central min-max optimization problems in ML, such as generative adversarial networks (GANs), robust reinforcement learning (RL), and distributionally robust ML, contain spurious attractors that do not include any stationary points of the original learning formulation. Indeed, we will describe how algorithms are subject to a grander challenge, including unavoidable convergence failures, which could explain the stagnation in their progress despite the impressive earlier demonstrations.