Contact

Fabian Sinz

Adresse: Spemannstr. 41
72076 Tübingen
Raum Nummer: 1.B.05

 

Bild von Sinz, Fabian

Fabian Sinz

Position: Doktorand  Abteilung: Alumni Bethge

I moved to the Lab of Prof. Jan Benda. Lastest information, code, notes, and publication can be found on my webpage.

Current Projects:

Picture

Natural Image Statistics We us the family of Lp-spherically and Lp-nested symmetric distributions to obtain more accurate morels of natural images and quantitatively assess normative hypotheses about the role of orientation selectivity and divisive normalization in the primate early visual system.

Further Reading

Past Projects:

Picture

UNIVERSVM: A SVM Implementation for Large Scale Transduction and Inference with a Universum The UNIVERSVM is a SVM implementation written in C++. Its functionality comprises large scale transduction (as described in Large Scale Transductive SVMs), sparse solutions (as described in Trading Convexity for Scalability) and inference with a universum (as described in Inference with the Universum).

  • Read online help
  • You can download Source code (C++ implementation) at mloss.org (just search for UniverSVM).

Picture

NEC-Labs ABCDetc letter/digit/symbol dataset (in collaboration Ronan Collobert and Jason Weston at NEC Labs America and Seyda Ertekin at PennState University)

We started to collect a dataset consisting of digits, symbols, uppercase and lowercase letters. At the moment it comprises around 50.000 examples.

If you want to contribute, please download the template , fill out the columns as indicated, scan it at 300 dpi and email it to me. (download page)


Transduction Loss as Linear Combination of Ramp Losses

Large Scale Optimization (in collaboration with Ronan Collobert, Jason Weston and Leon Bottou at NEC Labs America)

We show how the Concave-Convex Procedure can be applied to Transductive SVMs, which traditionally requires solving a combinatorial search problem. This provides for the first time a highly scalable algorithm in the nonlinear case.

(project details).


Grafik

Learning Depth from Stereo (Student Research Project) The depth of a point in space can be estimated by observing its image position from two different viewpoints. The classical approach to stereo vision calculates depth from the two projection equations which together form a stereocamera model. An unavoidable preparatory work for this approach is to estimate the parameters of the camera. This can become quite tedious.

In this study, we approached the depth estimation problem from a different point of view by applying generic machine learning algorithms to learn the mapping from image coordinates to spatial position. (project details).

Curriculum Vitae

  • October 2010 -December 2010: intership in the group of Gilles Laurent at the Max Planck Insitute for Brain Research in Frankfurt a.M.
  • March 2007 - January 2012: PhD student in the group of Matthias Bethge
  • August 2005 - October 2005: internship at the NEC laboratories in Princeton
  • 2002-2007: additional studies in philosophy
  • 2000-2007: studies in bioinformatics at the University of Tübingen

Employment

Scholarships and Awards

  • Best Paper Award at the Iternational Conference for Machine Learning 2006 (ICML 2006) for the paper Trading Convexity for Scalability
  • German National Academic Foundation (Studienstiftung des dt. Volkes, January 2008 - February 2010)

Teaching

  • Essential Mathematics for Neuroscience Lecture (Winterterm 2009, with J.-P. Lies), Graduate School of Neural and Behavioural Sciences, University of Tübingen
  • Essential Mathematics for Neuroscience Lecture (Winterterm 2008, with J. Macke), Graduate School of Neural and Behavioural Sciences, University of Tübingen
  • Essential Mathematics for Neuroscience Lecture (Winterterm 2007, with J. Macke), Graduate School of Neural and Behavioural Sciences, University of Tübingen
  • Ethics for Computer Scientists Seminar (Winterterm 2006, with P. Berens and D. Gümbel), University of Tübingen
  • Machine Learning and Neuroscience Practical Course (Winterterm 2004, with A. Gretton, Jeremy Hill and Dilan Görür), University of Tübingen

Präferenzen: 
Referenzen pro Seite: Jahr: Medium:

  
Zeige Zusammenfassung

Artikel (11):

Jiang X, Shen S, Cadwell CR, Berens P, Sinz F, Ecker AS, Patel S und Tolias AS (November-2015) Principles of connectivity among morphologically defined cell types in adult neocortex Science 350(6264) 1055: 1-10.
Sinz FH, Lies J-P, Gerwinn S und Bethge M (November-2014) Natter: A Python Natural Image Statistics Toolbox Journal of Statistical Software 61(5) 1-34.
Froudarakis E, Berens P, Ecker AS, Cotton RJ, Sinz FH, Yatsenko D, Saggau P, Bethge M und Tolias AS (Juni-2014) Population code in mouse V1 facilitates readout of natural scenes through increased sparseness Nature Neuroscience 17(6) 851–857.
Sinz F und Bethge M (November-2013) What Is the Limit of Redundancy Reduction with Divisive Normalization? Neural Computation 25(11) 2809-2814.
Sinz F und Bethge M (Januar-2013) Temporal Adaptation Enhances Efficient Contrast Gain Control on Natural Images PLoS Computational Biology 9(1) 1-13.
Theis L, Gerwinn S, Sinz F und Bethge M (November-2011) In All Likelihood, Deep Belief Is Not Enough Journal of Machine Learning Research 12 3071-3096.
Sinz F und Bethge M (Dezember-2010) Lp-Nested Symmetric Distributions Journal of Machine Learning Research 11 3409-3451.
pdf
Hosseini R, Sinz FH und Bethge M (Oktober-2010) Lower bounds on the redundancy of natural images Vision Research 50(22) 2213-2222.
pdf
Sinz FH, Gerwinn S und Bethge M (Mai-2009) Characterization of the p-Generalized Normal Distribution Journal of Multivariate Analysis 100(5) 817-820.
Eichhorn J, Sinz FH und Bethge M (April-2009) Natural Image Coding in V1: How Much Use is Orientation Selectivity? PLoS Computational Biology 5(4) 1-16.
Collobert R, Sinz F, Weston J und Bottou L (August-2006) Large Scale Transductive SVMs Journal of Machine Learning Research 7 1687-1712.
pdf

Beiträge zu Tagungsbänden (8):

Sinz F, Simoncelli EP und Bethge M (April-2010) Hierarchical Modeling of Local Image Features through Lp-Nested Symmetric Distributions In: Advances in Neural Information Processing Systems 22, , 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009), Curran, Red Hook, NY, USA, 1696-1704.
pdfpdf
Sinz F und Bethge M (Juni-2009) The Conjoint Effect of Divisive Normalization and Orientation Selectivity on Redundancy Reduction In: Advances in neural information processing systems 21, , Twenty-Second Annual Conference on Neural Information Processing Systems (NIPS 2008), Curran, Red Hook, NY, USA, 1521-1528.
pdf
Sinz FH, Chapelle O, Agarwal A und Schölkopf B (September-2008) An Analysis of Inference with the Universum In: Advances in neural information processing systems 20, , Twenty-First Annual Conference on Neural Information Processing Systems (NIPS 2007), Curran, Red Hook, NY, USA, 1369-1376.
pdf
Weston J, Collobert R, Sinz F, Bottou L und Vapnik V (Juni-2006) Inference with the Universum, 23rd International Conference on Machine Learning (ICML 2006), ACM Press, New York, NY, USA, 1009-1016.
pdf
Collobert R, Sinz F, Weston J und Bottou L (Juni-2006) Trading Convexity for Scalability, 23rd International Conference on Machine Learning (ICML 2006), ACM Press, New York, NY, USA, 201-208.
pdf
Quinonero Candela J, Rasmussen CE, Sinz F, Bousquet O und Schölkopf B (April-2006) Evaluating Predictive Uncertainty Challenge In: Machine Learning Challenges: Evaluating Predictive Uncertainty, Visual Object Classification, and Recognising Tectual Entailment, , First PASCAL Machine Learning Challenges Workshop (MLCW 2005), Springer, Berlin, Germany, 1-27, Series: Lecture Notes in Computer Science ; 3944.
pdf
Sinz F, Candela JQ, BakIr G, Rasmussen CE und Franz M (September-2004) Learning Depth From Stereo In: Pattern Recognition, , 26th Annual Symposium of the German Association for Pattern Recognition (DAGM 2004), Springer, Berlin, Germany, 245-252, Series: Lecture Notes in Computer Science ; 3175.
Görür D, Rasmussen CE, Tolias AS, Sinz F und Logothetis NK (September-2004) Modelling Spikes with Mixtures of Factor Analysers In: Pattern Recognition, , 26th Annual Symposium of the German Association for Pattern Recognition (DAGM 2004), Springer, Berlin, Germany, 391-398, Series: Lecture Notes in Computer Science ; 3175.
pdf

Beiträge zu Büchern (1):

Collobert R, Sinz F, Weston J und Bottou L: Trading Convexity for Scalability, 275-299. In: Large Scale Kernel Machines, (Ed) L. Bottou, MIT Press, Cambridge, MA, USA, (September-2007).
pdf

Technische Berichte (3):

Sinz FH und Bethge M: How Much Can Orientation Selectivity and Contrast Gain Control Reduce the Redundancies in Natural Images, 169, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, (März-2008).
pdf
Sinz FH und Schölkopf B: Minimal Logical Constraint Covering Sets, 155, Max-Planck Institute for Biological Cybernetics, Tübingen, Germany, (Dezember-2006).
pdf
Sinz FH: Kamerakalibrierung und Tiefenschätzung: Ein Vergleich von klassischer Bündelblockausgleichung und statistischen Lernalgorithmen, Wilhelm-Schickard-Institut für Informatik, Universität Tübingen, Tübingen, Germany, (März-2004).
pdf

Poster (10):

Cadwell CR, Jiang X, Sinz FH, Berens P, Fahey PG, Yatsenko D, Froudarakis E, Ecker AS, Cotton RJ und Tolias AS (Juni-2016): Cell Lineage Directs teh Precise Assembly of Excitatory Neocortical Circuits, AREADNE 2016: Research in Encoding And Decoding of Neural Ensembles, Santorini, Greece.
Reimer J, Yatsenko D, Ecker A, Walker EY, Sinz F, Berens P, Hoenselaar A, Cotton RJ, Siapas AG und Tolias AS (Juni-2016): DataJoint: Managing Big Scientific Data Using Matlab or Python, AREADNE 2016: Research in Encoding And Decoding of Neural Ensembles, Santorini, Greece.
Froudarakis A, Berens P, Ecker AS, Cotton RJ, Sinz FH, Yatsenko D, Saggau P, Bethge M und Tolias AS (Juni-2014): Population Code in Mouse V1 Facilities Read-out of Natural Scenes through Increased Sparseness, AREADNE 2014: Research in Encoding and Decoding of Neural Ensembles, Santorini, Greece.
Sinz F und Bethge M (September-13-2012): Temporal adaptation enhances efficient contrast gain control on natural images, Bernstein Conference 2012, München, Germany, Frontiers in Computational Neuroscience, Conference Abstract: Bernstein Conference 2012 67-68.
Theis L, Gerwinn S, Sinz F und Bethge M (Oktober-2010): Likelihood Estimation in Deep Belief Networks, Bernstein Conference on Computational Neuroscience (BCCN 2010), Berlin, Germany, Frontiers in Computational Neuroscience, 2010(Conference Abstract: Bernstein Conference on Computational Neuroscience).
Hosseini R, Sinz F und Bethge M (Oktober-2010): New Estimate for the Redundancy of Natural Images, Bernstein Conference on Computational Neuroscience (BCCN 2010), Berlin, Germany, Frontiers in Computational Neuroscience, 2010(Conference Abstract: Bernstein Conference on Computational Neuroscience).
Sinz F und Bethge M (September-30-2009): A new class of distributions for natural images generalizing independent subspace analysis, Bernstein Conference on Computational Neuroscience (BCCN 2009), Frankfurt a.M., Germany, Frontiers in Computational Neuroscience, 2009(Conference Abstract: Bernstein Conference on Computational Neuroscience) 114-115.
Sinz FH und Bethge M (Oktober-2008): The Conjoint Effect of Divisive Normalization and Orientation Selectivity on Redundancy Reduction in Natural Images, Bernstein Symposium 2008, München, Germany, Frontiers in Computational Neuroscience, 2008(Conference Abstract: Bernstein Symposium 2008).
Sinz F und Bethge M (Juli-29-2008): Redundancy Reduction in Natural Images: Quantifying the Effect of Orientation Selectivity and Contrast Gain Control, Gordon Research Conference: Sensory Coding & The Natural Environment 2008, Lucca, Italy.
Sinz F und Franz MO (Februar-2004): Learning Depth, 7th Tübingen Perception Conference (TWK 2004), Tübingen, Germany.
pdf

Abschlussarbeiten (1):

Sinz FH: A priori Knowledge from Non-Examples, Eberhard-Karls-Universität Tübingen, Germany, (März-2007). Diplom thesis
pdf

Vorträge (1):

Sinz F (Januar-27-2010) Invited Lecture: Contrast Gain Control in Natural Image Representations, UCL Gatsby Computational Neuroscience Unit, London, UK.

Export als:
BibTeX, XML, pubman, Edoc, RTF
Last updated: Montag, 22.05.2017