[cs-talks] Upcoming CS Seminars: Xiaokang Qiu of MIT (today) + PhD Proposal (Tues) + BUSec (Weds)
Conroy, Nora Mairead
conroynm at bu.edu
Thu Feb 12 07:24:09 EST 2015
Synthesizing Data-Structure Manipulations with Natural Proofs
Xiaokang Qiu, MIT
Thursday, February 12, 2015 at 3:30pm in MCS B29
Talk hosted by Hongwei Xi
Abstract: We presents an automatic approach to the synthesis of provably-correct implementations of data-structure manipulation routines. The input to our system is a program template together with a specification that the synthesized program must meet. Our system produces a program implementation along with necessary loop invariants and ranking functions, and guarantees the total correctness with a proof. Our approach is to reduce the synthesis problem over unbounded data-structures to a synthesis problem over a bounded domain by leveraging natural proof strategy.
To evaluate our approach we synthesize provably-correct manipulations on both standard list/tree data-structures and custom data-structures from OS kernels. Experiments show that our system can efficiently produce verified implementations for these data-structures in a matter of minutes.
Learning Space-Time Structures for Human Action Recognition and Localization
Shugao Ma, BU
Tuesday, February 17, 2015 at 4pm in MCS 148
Abstract: In this thesis we study the problem of action recognition and localization in realistic video clips. In training only the action class labels of the training video clips are available, and in testing both the action label and the spatial location of the action performer(s) are to be predicted. Although many past works have been done, this remains a challenging problem due to the complexity of the human actions and the large intra-class variations. Human actions are inherently structured patterns of body movements. However, past works are inadequate in learning the space-time structures in human actions and leveraging them for action recognition. In this thesis we propose new methods that exploit such space-time structures for effective human action recognition and localization. In the feasibility study, we developed a new local space-time representation for action recognition and localization, the hierarchical Space-Time Segments . Using this new representation, we explored ensembles of hierarchical spatio-temporal trees, discovered directly from training data, to model these structures for action recognition. This proposed approach in the feasibility study achieved state-of-the-art performances on two challenging benchmark datasets UCF-Sports and HighFive. However, further works are needed to more efficiently discover space-time structures and better handle large scale data. In the remaining work, we will explore deep convolutional neural network (CNN) for larger scale action recognition problem, studying ensemble of CNN models trained on whole video frames blended with motion information and CNN models trained on automatically proposed foreground regions. We will also explore sub-graph vectorization method that can effectively encode space-time structures of human actions into vectors, which will enable us to efficiently discover discriminative structures in a vector space. We will evaluate the remaining works on larger scale dataset, e.g. , the UCF101 dataset that has 101 action classes and 13000 videos.
Machine Learning Classification over Encrypted Data
Raphael Bost, DGA MI/Université de Rennes 1
Wednesday, February 18, 2015 at 9:30am in MCS 180 – Hariri Institute
Abstract: Machine learning classification is used in numerous settings nowadays, such as medical or genomics predictions, spam detection, face recognition, and financial predictions. Due to privacy concerns, in some of these applications, it is important that the data and the classifier remain confidential.
In this work, we construct three major classification protocols that satisfy this privacy constraint: hyperplane decision, Naïve Bayes, and decision trees. We also enable these protocols to be combined. At the basis of these constructions is a new library of building blocks for constructing classifiers securely; we demonstrate that this library can be used to construct other classifiers as well, such as a multiplexer and a face detection classifier.
We implemented and evaluated our library and classifiers. Our protocols are efficient, taking milliseconds to a few seconds to perform a classification when running on real medical datasets.
Joint work with Raluca Ada Popa, Stephen Tu and Shafi Goldwasser which will appear in NDSS'15.
Computational Understanding of Image Memorability
Zoya Bylinskii, MIT
Thursday, February 19, 2015 at 4pm in MCS 148
Abstract: In this talk, I will describe the research done in the Oliva Lab on Image Memorability - a quantifiable property of images that can be used to predict whether an image will be remembered or forgotten. Apart from presenting the lab's research directions and findings, I will focus on the work I have done in understanding and modeling the intrinsic and extrinsic factors that affect image memorability. I will present results on how consistent people are in which images they find memorable and forgettable (across experiments, settings, and visual stimuli) and I will show how these findings generalize to information visualizations. I will also demonstrate how the extrinsic factors of image context and observer eye behavior modulate image memorability. I will present an information-theoretic model of context and image distinctiveness, to quantify their effects on memorability. Finally, I will demonstrate how eye movements, pupil dilations, and blinks can be predictive of image memorability. In particular, our computational model can use an observer's eye movements on an image to predict whether or not the image will be later remembered. In this talk, I hope to offer a more complete picture of image memorability, including the contributions to cognitive science, and the computational applications made possible.
The following is the first paper on image memorability that has come out of the Oliva Lab, and has started a whole direction of research: http://cvcl.mit.edu/papers/IsolaXiaoTorralbaOliva-PredictingImageMemory-CVPR2011.pdf -- it can give people some background, though I will provide an intro as well.
Bio: Zoya Bylinskii is a PhD student at MIT, jointly supervised by Aude Oliva and Fredo Durand. She works in the area of computational perception - at the intersection of cognitive science and computer science. Specifically, she is interested in studying human memory and attention, in order to build computational models to advance the understanding and application possibilities of these areas. Her current work spans a number of research directions, including: image memorability, saliency benchmarking, and information visualizations. Zoya most recently completed her MS under the supervision of Antonio Torralba and Aude Oliva, on a "Computational Understanding of Image Memorability". Prior to this, her BS research on parts-based object recognition was supervised by Sven Dickinson at the University of Toronto. She also spent a lovely summer in 2011 working in BU with Stan Sclaroff on reduplication detection in sign language :)
Improving Face Analysis Using Expression Dynamics
Monday, February 23, 2015 at 3pm in MCS 148
Abstract: Most of the approaches in face analysis rely solely on static appearance. However, temporal analysis of expressions reveals interesting patterns. In this talk, I will describe automatic spontaneity detection for enjoyment smiles using temporal dynamics of different facial regions. We have recorded spontaneous and posed enjoyment smiles of hundreds of visitors to the NEMO Science Centre in Amsterdam, thus creating the most comprehensive smile database ever: the UvA-NEMO Smile Database (www.uva-nemo.org). Our findings on this publicly available database show that facial dynamics go beyond expression analysis. I will discuss how we can use expression dynamics to improve age estimation and kinship detection.
Bio: Hamdi Dibeklioglu received the B.Sc. degree from Yeditepe University, Istanbul, Turkey, in 2006, the M.Sc. degree from Bogazici University, Istanbul, Turkey, in 2008, and the Ph.D. degree from the University of Amsterdam, Amsterdam, The Netherlands, in 2014. He is currently a Post-Doctoral Researcher with the Pattern Recognition and Bioinformatics Group, Delft University of Technology, Delft, The Netherlands. He is also a Guest Researcher with the Intelligent Systems Lab Amsterdam, University of Amsterdam. His research interests include computer vision, pattern recognition, and automatic analysis of human behavior.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the cs-talks