Flower Machine Learning Group
University of Hawaii at Manoa, ICS Department

PI: Susanne Still
 Flower

PhD Students
Lisa Miller picture Projects:

|| Robotics
Lisa's passion is robotics. She is currently working on new algorithms for exploration and self organized behavior.

|| Document Classification
Lisa is creating a tool for astrobiologists to faster access relevant papers from different areas. A part of this problem is to classify the documents. Lisa is currently testing various different classification techniques and a new method we have developed for document preprocessing.


William Wright		 Bill Wright Picture Projects:

|| Clustering Algorithms
William is working on issues of robustness and efficiency.

|| Whale Song Analysis
William's true passion is the analysis of Whale Songs. He will be applying the clustering methods he develops to this problem.


Emiliano Miranda		      picture

Project:

|| Games and Psychophysics
Emiliano has a background working in the computer game industry. He is exploring the use of games in psychophysical experiments. He is interested in questions related to how humans learn, and how they make decisions.

Lane McIntosh
(Masters student, UHM Mathematics)

LanePhoto

 Project:

|| Information Processing and Energy Dissipation in Neurons
Neurons are computing machines that are designed to learn, i.e. they memorize some aspects of the past. We have shown recently (Still et al. 2012) that under fixed memory, the maximization of predictive power results in efficient computation, in the sense that dissipation is minimized. Lane is investigating the consequences of this insight on neuronal adaptation.

Affiliated former students
|| Dr. Christopher Hamilton, Geophysics, UHM, now at NASA.
|| Dr. Hunter Hatfield, Linguistics, UHM, now at the University of Otago.
|| Dr. Fabio Caccioli, SISSA and SFI
|| Dr. Christopher Ellison, Physics, UC Davis
|| Monica Dinculescu, CS, McGill
|| Victor Miagkikh, ICS, UHM
Flower Lab Meeting

Tue 3.30 - 4.30 pm
P.O.S.T. 302.		 .
If you are interested to learn more about what is going on in our lab, you are very welcome to join the lab meeting. Please send an email to announce your attendance.

Information for prospective students: Research interests and open projects

Current Collaborations (incomplete list, under construction)
Research Area  (Co)-PIs
  Outside Group Members
ML Group Students
 Publications
Nonequilibrium Thermodynamics and Physics of Computation
 || Gavin Crooks, LBNL and UC Berkeley
|| Tony Bell, RCTN, UC Berkeley
 || David Sivak, LBNL and UC Berkeley Looking for a Physics Student
 S. Still, D. A. Sivak, A. J. Bell, and G. E. Crooks. The thermodynamics of prediction.
Quantitative Finance / Econophysics
 ||Imre Kondor, Collegium Budapest
|| Matteo Marsili, ICTP		 || Fabio Caccioli, SFI
|| Zeno Farkas
 Looking for a CS, physics, mathematics or economics student with strong analytical and programming skills
 S. Still and I. Kondor. Regularizing Portfolio Optimization.  New Journal of Physics 12 (2010) 075034 (15pp). Special Issue on Statistical Physics Modeling in Economics and Finance. This work was also presented at the Third Financial Risks International Forum, Paris, March 25/26, 2010.

F. Caccioli, S. Still, M. Marsili, and I. Kondor: Optimal Liquidation Strategies Regularize Portfolio Selection. To be published in The European Journal of Finance, Special issue on New Facets of Economic Complexity in Modern Financial Markets (2011).
Robotics, information theory, time series analysis
 || Bill Bialek, Physics, Princeton
|| Jim Crutchfield, Physics, UC Davis
||Doina Precup, CS, McGill
|| Fritz Sommer, RCTN, UC Berkeley
|| Leon Bottou, Microsoft
 ||Chris Ellison, UC Davis
|| Benny Brown, UC Davis
|| Daniel Little, RCTN, UC Berkeley		 ||Lisa Miller
 S. Still. Information theoretic approach to interactive learning. EPL 85 (2009) 28005.

S. Still, J. P. Crutchfield and C. J. Ellison. Optimal Causal Inference: Estimating Stored Information and Approximating Causal Architecture. CHAOS 20 (2010) Special Issue on "Intrinsic and Designed Computation: Information Processing in Dynamical Systems".

S. Still and W. Bialek. How many clusters? An information theoretic perspective. Neural Computation, 16(12):2483-2506 (2004)  

S. Still, W. Bialek and Bottou, L. Geometric Clustering using the Information Bottleneck method. In Sebastian Thrun, Lawrence K. Saul, and Bernhard Schölkopf, editors, Advances in Neural Information Processing Systems 16. MIT Press (2004)

E. Schneidman, S. Still, M. J. Berry II, and W. Bialek. Network Information and Connected Correlations. Physical Review Letters, 91(23):238701 (2003)
Geoinformatics
 || Christopher Hamilton, NASA
||Thorvaldur Thordarson, U Edinburgh

||Bill Wright
Looking for a Geophysics Student
 C. W. Hamilton, C. D. Beggan, S. Still, M. Beuthe, R. Lopes, D. A. Williams, W. Wright and J. Radebaugh. Spatial distribution of volcanoes on Jupiter’s moon Io: implications for tidal heating and magma ascent. (Preprint available upon request.)
Foundations of Information Theory and Thermodynamics
 || Mike DeWeese, Physics UC Berkeley
|| Rob Shaw
 ||Badr Faisal Albanna, RCTN, UC Berkeley
|| Jascha Sohl-Dickstein, RCTN, UC Berkeley
 Looking for a Physics Student Coming up soon! Stay tuned.
Computational Neuroscience
 || Giacomo Indiveri, INI, ETH/UNI Zuerich ||Steimer, Andreas, INI, ETH/UNI Zuerich
 || Lane McIntosh
Applying Statistical Mechanics and Machine Learning to Climate Modeling
 ||Peter Mueller, SOEST, UHM
Looking for a Student in Oceanography


Machine learning research and related activities at UH:

Manoa Lecture Series in Machine Learning and Computational Neuroscience

Tony Kuh's Lab, Electrical Engineering

ICS Robotics Group (PIs)
Kyungim Baek (Computer Vision)
Kim Binsted (AI)
David Chin (AI)
Curtis Ikehara (HCI)
Nancy Reed (AI)
Susanne Still


ARE YOU WORKING IN MACHINE LEARNING AT UH AND ARE NOT ON THIS LIST? PLEASE GET IN TOUCH TO BE INCLUDED.