Benjamin Auffarth, PhD

About

I am interested in finding out what makes people tick. I hold PhD and MSc titles in machine learning in neuroscience, and I have research experience at different levels always in multi-disciplinary environments. I've worked on data algorithms, neuroscience, psychology, and databases. I've conducted and analyzed experiments with data of >3TB. I've run brain models of several hundred thousand neurons on IBM supercomputers. I've designed and conducted experiments on the brain. 

I have been working with quantitative analyses for several years, culminating in a PhD degree from the Royal Institute of Technology in Stockholm, Sweden in 2012. Additionally, among other things, I am fluent in several programming and scripting languages, and have implemented websites, submission platforms for university courses, built linux clusters, and implemented models of the brain on high-performance computers. 

Biosketch
Growing up between Mannheim and Heidelberg, on a long-term quest for entropy reduction, Benjamin has been leading a peripatetic life. He has been living in different places including St. Petersburg, Sofia, Ankara, Tokyo, Barcelona, Stockholm, Berlin, and Heraklion, soaking up many pieces of information and acquiring formal education in different subjects including computer vision, artificial intelligence, computational complexity, neural network theory, biology, neuroscience, cognitive science, linear algebra, probability and optimization theory. After finishing his PhD at the KTH Royal Institute of Technology with Professors Anders Lansner and Tony Lindeberg, and a research stay in Berlin at the Freie Universitaet Berlin, he is currently working as a Postdoctoral Fellow in the Computational Biology Laboratory at the Institute of Molecular Biology and Biotechnology (IMBB) of the Foundation of Research and Technology-Hellas (F.O.R.T.H.) in Heraklion, Greece.

The main contributions of his thesis are about the early olfactory system on aspects of odor representations and understanding the stimulus. How odor stimuli are mapped spatially in the olfactory system has been controversial. Studies conducted in the course of his Ph.D. work confirmed and extended findings about a spatial decomposition clustering of representative areas) of odorants by several physico-chemical properties of odorant molecules. Another study suggested a spatial decomposition by perceptual properties. A third study investigated hypotheses of how a spatial ordering could comes about from principles of sensory axons guidance, which were simulated in an abstract model of the early olfactory system. Mechanisms included a plausible distribution of receptor response profiles to a set of biologically relevant odorants, further sensory neurons, and the sensory convergence targets called glomeruli. This study showed that a sensory axon convergence by activity dependence ("fire-together-wire-together") could give an account of several phenomena.

Before working on the olfactory system, he finished his MSc degree in the group for Volume Visualization and Artificial Intelligence (WAI) at the Universitat de Barcelona (UB. He studied machine learning techniques and graduated with a thesis on feature selection in biomedical images, and published findings. At the same time he worked on a project to develop an assignment management system (baptized: S.A.C.) for university programming classes. 

He holds a B.Sc. diploma in Cognitive Science from the University of Osnabrueck, a degree which he supplemented with studies at the New Bulgarian University (NBU), the Middle East Technical University (METU), and the Chuo University in Tokyo. Overall, he took many optional subjects on computer science, psychology, computational linguistics, and machine learning, and graduated with a thesis comparing eye traces of rhesus macaque monkeys to those of human subjects (using machine learning and statistics). 

Research topics

How the mind and its substrate, the brain, works is a challenging problem. Ultimately, how the nervous system really works will depend on understanding its structure, including the circuitry between cells, and its physiology. I am interested in investigating theories and producing testable models with plausible assumptions and useful abstractions. 

The brain is the substrate of information processing and memory. It is an information processing device whose power depends on dendritic complexity, the activation of ionic channels and on a synaptic network between neurons and their neuritic structures. Learning and memory are cognitive phenomena that depend on the health of the brain, however, the it is hard to relate very abstract processes to changes at the microscopic scale. It is the central tenet of neuroscience that neurons are the principal processing elements of the brain, and, in recent years it has become increasingly clear that powerful mechanisms at the synaptic level influence processing, and make neurons and therefore the brain much more powerful than previously believed. It is part of our work to make the bridge between cognitive processes at the network level and key elements on the microscopic scale using functional and structural measures in order to facilitate understanding of the brain. This is basic science, however has bearing on medical applications and the production of neuromorphic devices. We pursue modeling studies, where we investigate the network formation between neurons, and how the electric coupling between dendritic branches and the soma can change based on their history of excitation. These models are being validated by comparison to work in neurophysiology and neuroimaging. Further, we work in collaborations with other labs to conduct neurophysiological and neuroimaging studies on cortical and hippocampal networks, to study the organization and structure of learning and memory. 

Past and current research topics include:

  • dendritic processing, 
  • plasticity and memory, 
  • genetic algorithms, 
  • the early olfactory system, 
  • modeling of biological systems, 
  • machine learning and data mining, 
  • cell tracking, 
  • feature selection, 
  • artificial fixations (artificial scanpath), and
  • eye movements (eye tracking). 
I've worked with with statistical analyses and machine learning, computational modeling at different levels of complexity and scales, and experimental techniques such as multi-electrode arrays, patch-clamping, and eye-tracking. 

Publications

Journals

Benjamin Auffarth. “Understanding smell - the olfactory stimulus problem.” Neuroscience & Biobehavioral Reviews, 37 (8), pp 1667-1679. [article link]

Benjamin Auffarth, Bernhard Kaplan, Anders Lansner. "Map formation in the olfactory bulb by axon guidance of olfactory neurons." Frontiers in Systems Neuroscience5(0), 1-16, 2011. doi:10.3389/fnsys.2011.00084 [open access article]

Benjamin Auffarth, Gutierrez-Galvez Galvez, Santiago Marco. "Continuous spatial representations in the olfactory bulb may reflect perceptual categories." Frontiers in Systems Neuroscience5(0), 1-14, 2011. doi:10.3389/fnsys.2011.00082 [open access article]

Benjamin Auffarth, Agustín Gutierrez-Galvez, Santiago Marco. "Statistical analysis of coding for molecular properties in the olfactory bulb." Frontiers in systems neuroscience, 5(0), 62, 2011. doi:10.3389/fnsys.2011.00062 [open access article]

Book chapters

Benjamin Auffarth, Maite López-Sánchez, Jesús Cerquides. "Comparison of Relevance and Redundancy Measures for Feature Selection." P. Perner (Ed.), Advances in Data Mining - Applications in Medicine, Web Mining, Marketing, Image and Signal Mining, lnai 6171, Springer-Verlag, Heidelberg, 2010, pp. 248-262.doi:10.1007/978-3-642-14400-4_20. [pdf]

Benjamin Auffarth, Maite López-Sánchez, Jesús Cerquides. "Hopfield Networks in Relevance and Redundancy Feature Selection Applied to Classification of Biomedical High-Resolution Micro-CT Images." P. Perner (Ed.), Advances in data mining: medical applications, e-commerce, marketing, and theoretical aspects. LNAI 5077, Springer-Verlag, Heidelberg, 2008, pp 16-31.doi:10.1007/978-3-540-70720-2_2. [pdf]

Conference papers

Benjamin Auffarth. “Clustering by a genetic algorithm with biased mutation operator.“ 2010 IEEE WCCI CEC. Barcelona, Spain, July 18–23, 2010.doi:10.1109/CEC.2010.5586090. [pdf]

Benjamin Auffarth, Agustín Gutierrez, Santiago Marco. “Relevance and Loci of Odorant Features in the Rat Olfactory Bulb.“A. L. N. Fred, J. Filipe, & H. Gamboa (Eds.), BIOSIGNALS 2010 - Proceedings of the Third International Conference on Bio-inspired Systems and Signal Processing (pp. 37-44). [pdf]

Matteo Falasconi, Agustín Gutierrez, Benjamin Auffarth, Giorgio Sberveglieri, Santiago Marco. "Cluster Analysis of the Rat Olfactory Bulb Activity in Response to Different Odorants."M. Pardo & Giorgio Sberveglieri (Eds.), OLFACTION AND ELECTRONIC NOSE: Proceedings of the 13th International Symposium on Olfaction and Electronic Nose (Vol. 1137, pp. 169-172). Brescia (Italy): AIP. [pdf]

Benjamin Auffarth, Maite López-Sánchez, Jordi Campos, Anna Puig. "System for Automated Assistance in Correction of Programming Exercises (SAC)", 5th international congress of university teaching and innovation (CIDUI). Lleida, Spain. July 2-4th, 2008. pp. 104 (1-9). [pdf]

Benjamin Auffarth, Yasumasa Mutou and Yasuharu Kunii. "An Artificial System for Visual Perception in Autonomous Robots." Proceedings of the 9th International Conference on Intelligent Engineering Systems, 2005.Auffarth, B., Muto, Y., & Kunii, Y. (2005). doi:10.1109/INES.2005.1555159. 

Others

Benjamin Auffarth, "How to build a beowulf," Technical report, November 2008 [pdf]

Jordi Fonollosa, Benjamin Auffarth. "Sensores de gas y narices electrónicas." GEA, 61, pp. 52-53, 2008.

S. Napagao, B. Auffarth, N. Ramirez. "Agent Language Analysis: 3-APL." Technical report. UPC, 2007.

B. Auffarth. "Spectral graph clustering." Technical report. UPC, 2007. 

Posters

Benjamin Auffarth, Dirk Van Dyck, Michele Giugliano, Panagiota Poirazi. "Burst synchronization in circuits of cortical neurons in a closed-loop stimulation." presented at Eninet 2014, Freiburg (Breisgau). September 2014.

Benjamin Auffarth, Panagiota Poirazi. "Late LTP and heterosynaptic effects in a biophysical model of synaptic tagging and capture." presented at Dendrites 2014, Heraklion, Greece. July 2014.

Benjamin Auffarth, Panagiota Poirazi. "Plastic A-type potassium channels in a detailed morphological neuron model." SN/ FENS School 2013 on Local protein synthesis in axons and dendrites, 24-28 September 2013, Orthodox Academy of Crete, Kolymbari, Crete, Greece. 

Benjamin Auffarth. "Large-scale analysis of perceptual odor data." presented at Odorspaces 2013, Hannover; published in Flavour, 3(Suppl 1), P1, 2014.

Benjamin Auffarth. "Understanding physico-chemical, physiological, and percepual aspects of the olfactory stimulus - a computational perspective on olfaction." 8th FENS Forum of Neuroscience, July 14-18, 2012.

Benjamin Auffarth. "Activity-dependent memory organization in the early mammalian olfactory pathway for decorrelation, noise reduction, and sparseness-enhancement."BMC Neuroscience (Vol. 12, p. P186). BioMed Central Ltd. doi:10.1186/1471-2202-12-S1-P186 [abstract]

Benjamin Auffarth, Agustín Gutiérrez-Galvez, Santiago Marco. "Which Odorant Properties Are Most Relevant to Perception?" Future Emerging Technologies (FET). Prague, April 21–23, 2009.

Benjamin Auffarth, Agustín Gutiérrez-Galvez, Santiago Marco. "Study of chemotopic representations of glomeruli in the rat olfactory bulb." ECRO 2008. Portoroz, Slovenia. September 2008.

Matteo Falasconi, Benjamin Auffarth, Agustín Gutiérrez-Galvez, Santiago Marco. "Exploring the number of modular dimensions in olfactory bulb activations of the rat." ECRO 2008. Portoroz, Slovenia. September 2008.

Yasumasa Mutou, Benjamin Auffarth, Yasuharu Kunii. "A Model of Human Vision for Application in Autonomous Robots." 2005. Robocon, Kobe.

Talks

"Investigations Into Odor Representations at the Rat Olfactory Bulb", PhD discussion seminar, IBEC, February 2009.

Theses

"Machine Learning Techniques with Specific Application to the Early Olfactory System." PhD thesis, KTH Royal Institute of Technology, Presented march 2012. [link]

"Feature Selection in Classification of Biomedical High-Resolution Micro-CT Images." Master Thesis, University of Barcelona, Presented December 2007. [pdf]

"An Investigation into Differences of Eye Movements in Monkeys and Humans." Bachelor Thesis, University of Osnabrück, Submitted September 2006.


Scholarships

Benjamin was being supported (at different times) by competitive scholarships including 

  1. Marie Curie ITN
  2. DAAD
  3. Socrates
  4. Monbukagakusho
  5. FI (Formació de personal investigador), and 
  6. ECRO


Contact

auffarth <nospam> at </nospam> imbb dot forth dot se 


Links

Research papers by Benjamin Auffarth

View Benjamin Auffarth's profile on LinkedIn

- interview with DAAD Athens(available in German and Greek)

- Researcher ID