Intelligent and Adaptive Systems Seminar


This is the schedule for the Intelligent and Adaptive Systems seminar at Portland State University. This seminar is an informal weekly meeting of students, faculty, and others on topics related to artificial intelligence, machine learning, cognitive science, and complex adaptive systems. The seminar is coordinated by Will Landecker. If you would like to attend meetings or subscribe to the the mailing list for this seminar, e-mail Will.

Academic Year 2011-2012

Spring 2012

Friday April 6, 2012 (3pm, FAB 155). Melanie and Mandar will present Direction estimation of pedestrian from multiple still images (Shimizu and Poggio) and Image based estimation of pedestrian orientation for improving path prediction (Gandhi and Trivedi).

Friday April 13, 2012 (3pm, FAB 155). Mick Thomure will present Natural image coding in V1: How much use is orientation selectivity? (Eichhorn, Sinz and Bethge, PLOS Comp Bio 5-4, 2009)

Friday April 20, 2012 (3pm, FAB 155). George Dittmar will present Using body-anchored priors for identifying actions in single images (Karlinsky, Dinerstein and Ullman, NIPS 2010).

Friday April 27, 2012 (3pm, FAB 155). Dona Hertel will present Recognizing Face or Object from a Single Image: Linear vs. Kernel Methods on 2D Patterns (Zhang, Chen and Zhou, SSPR 2006).

Friday May 4, 2012 (3pm, FAB 155). TBA.

Friday May 11, 2012 (3pm, FAB 155). TBA.

Winter 2012

Friday January 13, 2012 (3pm, FAB 155). Mick Thomure will present Learning to select features using their properties by E. Krupka, A. Navot, and N. Tishby (JMLR 2008)

Friday January 20, 2012 (3pm, FAB 155). Vitaliy Pavlenko will present Improved Annotation of the Blogosphere via Autotagging and Hierarchical Clustering by Brooks and Montanez (WWW 2006).

Friday January 27, 2012 (3pm, FAB 155). Nathan Nifong will present Style-based inverse kinematics by Grochow et al. (SIGGRAPH 2004).

Friday February 3, 2012 (3pm, FAB 155). John Ledesma will present Support Vector Machines Under Adversarial Label Noise by Biggio, Nelson and Laskov (ACML 2011).

Friday February 10, 2012 (3pm, FAB 155). This week, we will watch a tutorial on Gaussian Processes, by David MacKay: http://videolectures.net/gpip06_mackay_gpb/

Friday February 17, 2012 (3pm, FAB 155). Max Quinn will present Learning Better Image Representations Using 'Flobject Analysis' (Li et al., CVPR 2011).

Friday February 24, 2012 (3pm, FAB 155). George Dittmar will present 3D Convolutional Neural Networks for Human Action Recognition (Ji et al., ICML 2010).

Friday March 2, 2012 (3pm, FAB 155). Mick Thomure will present Unbiased look at dataset bias (Torralba and Efros, CVPR 2011).

Friday March 9, 2012 (3pm, FAB 155). John Ledesma will present Robust feature extraction via information theoretic learning (Yuan and Hu, ICML 2010).

Friday March 16, 2012 (3pm, FAB 155). Kevin Dyer will present his research on traffic analysis countermeasures: Peek-a-boo, I still see you: why efficient traffic analysis countermeasures fail (Dyer et al., IEEE Symposium on Security & Privacy 2012).

For background reading on traffic analysis, see http://www.freehaven.net/anonbib/cache/morphing09.pdf and http://lorre.uni.lu/~andriy/papers/acmccs-wpes11-fingerprinting.pdf.

Fall 2011

Friday September 9, 2011 (3pm, FAB 155). Will Landecker will present Robust Classification of Objects, Faces, and Flowers Using Natural Image Statistics by Kanan and Cottrell (CVPR 2010).

Friday September 16, 2011 (3pm, FAB 155). Mick Thomure will present Are sparse representations really relevant for image classification? by Rigamonti, Brown and Lepetit (CVPR 2011).

Friday September 23, 2011 (3pm, FAB 155). Max Quinn will present Learning complex cell invariance from natural videos: A plausibility proof by Masquelier, Serre, Thorpe and Poggio (MIT Tech. Report, 2007).

Friday September 30, 2011 (3pm, FAB 155). Max OrHai will present his paper Spatial Sorting Algorithms for Parallel Computing in Networks, in preparation for his presentation at IEEE Conf. Self-Adaptive and Self-Organizing Systems (2011).

Friday October 7, 2011, (3pm, FAB 86-01). Vitaliy Pavlenko will present Discriminative Parameter Learning for Bayesian Networks by Su et al. (ICML 2008).

Friday October 14, 2011 (3pm, FAB 155). George Dittmar will present Active learning in multimedia annotation and retrieval: A survey by Wang and Hua (ACM TIST; Vol 2, Issue 2; 2011).

Friday October 21, 2011 (3pm, FAB 155). Alireza Goudarzi will present Connectivity, Dynamics, and Memory in Reservoir Computing with Binary and Analog Neurons by Busing et al. (Neural Computation, 2010).

Friday October 28, 2011 (3pm, FAB 155). Will LAndecker will present Independent Component Analysis: Algorithms and Applications by Hyvarinen and Oja (Neural Networks, 2000).

Friday December 16, 2011 (3pm, FAB 155). We will watch the lecture Rich Probabilistic Models for Holistic Scene Understanding by Daphne Koller (IJCAI 2011).

Academic Year 2010-2011

Friday March 4, 2011 (3pm, FAB 135-2). Will Landecker will present Deconvolutional Networks by Matthew D. Zeiler, Dilip Krishnan, Graham W. Taylor and Rob Fergus (CVPR 2010).

Friday March 11, 2011 (2pm, FAB 155). Mick Thomure will present Factorial coding of natural images: how effective are linear models in removing higher-order dependencies? by M. Bethge (Journal of the Optical Society of America, 2006).

Friday March 18, 2011. No meeting due to finals.

Friday March 25, 2011. No meeting due to spring break.

Friday April 8, 2011 (2:30pm, FAB 155). Will Landecker will present Learning mid-level features for recognition by Boureau et al. (CVPR 2010).

Friday April 15, 2011 (2:30pm, FAB 155). John Wehland will present "Co-Evolving Parasites Improve Simulated Evolution as an Optimization Procedure" by Daniel Hillis (Artificial Life, 1991).

Friday April 23, 2011 (2:30pm, FAB 155). Max Quinn will present A Bayesian model for efficient visual search and recognition by Elazary and Itti (Vision Research, 2010).

Friday April 30, 2011 (2:30pm, FAB 155). Martin Cenek will give a practice talk in preparation for his thesis defense.

Friday May 6, 2011 (2:30pm, FAB 155). Will Landecker will present Classification using intersection kernel support vector machines is efficient by Maji et al. (CVPR 2008) and Max-margin additive classifiers for detection by Maji and Berg (ICCV 2009).

Friday May 13, 2011. The CS department Research Profficiency Exam (RPE) takes place today. Instead of the usual IAS seminar, we suggest attending the RPE talks.

Friday May 20, 2011. Danny Voils will present Simultaneous tracking and activity recognition with relational dynamic Bayesian networks by Manfredotti et al. (U. Regina Tech Report, 2011).

Friday May 27, 2011

Friday June 3, 2011

The archives of the seminar can be found here.]