[Eecs-news-links] FW: UO CIS Colloquia, TUESDAY April 26, and THURSDAY, April 28, 2016 @ 3:30pm in 220 Deschutes Hall

Batten, Tina tina.batten at oregonstate.edu
Mon Apr 25 08:15:29 PDT 2016

-----Original Message-----
From: Adriane Bolliger [mailto:adriane at cs.uoregon.edu] 
Sent: Monday, April 25, 2016 8:15 AM
To: colloquia at cs.uoregon.edu; dept at cs.uoregon.edu
Cc: grads-mail at cs.uoregon.edu
Subject: UO CIS Colloquia, TUESDAY April 26, and THURSDAY, April 28, 2016 @ 3:30pm in 220 Deschutes Hall

Colloquium #1: TUESDAY, April 26

Lorenzo De Carli 
University of Wisconsin-Madison

Effective network security in the golden age of online threats


Intrusion prevention systems (IPSs), which analyze network traffic to detect signs of malicious activity, are a long-standing cornerstone of network security. Nowadays, the combination of advanced, targeted online threats and increasing bandwidth usage is making existing tools increasingly ineffective. In order to cope with the large amounts of data moved by network links, current IPSs limit themselves to simple threat detection strategies which match each network flow against a set of attack signatures. This approach is fragile and limited in expressiveness: signatures can be often evaded by small tweaks in the attack strategy, and fail to capture various classes of attacks altogether.

In my talk I will describe the design of a flexible IPS platform which supports complex threat detection strategies, while satisfying the performance requirement through parallelization. In particular, my work proposes a domain-specific concurrency model, in which a work scheduler partitions network traffic into subsets that can be analyzed independently for threat detection purposes. This scheduler drives a multi-threaded IPS in which concurrent threads always process independent slices of network traffic, making synchronization and inter-thread communication unnecessary. The system uses a novel program analysis technique to automatically generate a suitable work scheduler given any user-defined threat detection algorithm. This makes parallelization general and fully transparent to the operator.

In the second part of my talk I will provide an overview of another relevant contribution of my Ph.D. work: a programmable dataflow-based hardware accelerator for inspection and forwarding of network traffic.


Lorenzo De Carli is a Ph.D. candidate in Computer Science at the University of Wisconsin-Madison, advised by Somesh Jha. His research interests focus on networking and security, including intrusion prevention and packet processing. His contributions include parallelization strategies for intrusion prevention, hardware accelerator for packet inspection and forwarding, and analysis of malware communications. He has also worked on optimized signature matching and instruction scheduling for novel processor architectures. Lorenzo received a B.Sc. (2004) and a M.Sc. (2007) in Computer Engineering from Politecnico di Torino, Italy, and a M.Sc. in Computer Science (2010) from the University of Wisconsin-Madison.

DATE:	Tuesday, April 26, 2016 
TIME:	3:30 p.m. talk, refreshments following talk
PLACE:	220 Deschutes Hall (Colloquium Room), University of Oregon


Colloquium #2: THURSDAY, April 28

Samuel Gerber 
University of Oregon

Geometrical Approaches for Analyzing and Visualizing High Dimensional Data


High dimensional data arises in a variety of applications. In neurological studies, large data sets of diffusion tensor and magnetic resonance images, consisting of millions of measurements, are acquired. In climate science there is a growing demand to quantify the uncertainty in simulations, which are controlled by up to a hundred parameters. For scientists working with such data, it is often very difficult to gain a qualitative understanding to reason and hypothesize about the underlying process. Thus, data models that convey insights into structures present in the data are exceedingly important and are the focus of this talk.

In this talk I will discuss two techniques, each with respect to a particular application: (1) Population analysis from medical images and (2) parameter space exploration of climate simulations. For the first application I will focus on dimension reduction and manifold learning and introduce a new approach based on principal surfaces. For the second application I exploit ideas from computational topology for visualization and regression of high dimensional scalar functions.


Samuel Gerber is currently working at the University of Oregon providing support and doing research on computational and data analysis methodology. Before moving to Eugene he was a visiting assistant professor at Duke working with Prof. Mauro Maggioni. He finished his PhD in 2012 under the supervision of Prof. Ross Whitaker at the SCI Institute at the University of Utah.

His research interests span across machine learning, optimization and visualization. Of particular interest are geometric approaches to high dimensional data sets. His work on manifold models for brain population analysis won a best paper award at MICCAI 2010 and received a young scientist publication impact award at MICCAI 2014.

DATE:	Thursday, April 28, 2016 
TIME:	3:30 p.m. talk, refreshments following talk
PLACE:	220 Deschutes Hall (Colloquium Room), University of Oregon

For all CIS public talks, go to:

More information about the Eecs-news-links mailing list