A. Emin Orhan

Current affiliation: Center for Data Science, New York University E-mail: [email protected] Alternative e-mail: [email protected] Web page: https://sites.google.com/view/eminorhan Google Scholar: https://scholar.google.com/citations?user=YkT8jLoAAAAJ&hl=en Research blog: https://severelytheoretical.wordpress.com/ GitHub: https://github.com/eminorhan

Research Areas

Computational neuroscience, cognitive science, machine learning.

Education

– Post-doc, Center for Data Science, New York University, January 2019 - current.

Position jointly funded by Dr. Brenden Lake and by a Moore-Sloan Fellowship.

– Post-doc, Department of Neuroscience, Baylor College of Medicine, September 2016 - January 2019.

Affiliated with the Department of Electrical & Computer Engineering at Rice University through my advisor Advisor: Dr. Xaq Pitkow

– Post-doc, Center for Neural Science, New York University, September 2013 - September 2016.

Advisor: Dr. Wei Ji Ma

– Ph.D. Brain & Cognitive Sciences, University of Rochester, September 2008 - August 2013.

Thesis: Theoretical Investigations in Visual Short-Term Memory: Structured Probabilistic Representations, Model Mismatch and Neural Population Coding Advisor: Prof. Robert A. Jacobs M.A. obtained in October 2011

– M.A. Cognitive Science, Bogazici University, September 2005 - June 2008.

Thesis: Unsupervised Learning of High-Level, Invariant Visual Representations through Temporal Coherence Advisor: Prof. Ethem Alpaydın

– B.A. Philosophy (with Highest Honors), Bogazici University, September 2002 - August 2005.

A. Emin Orhan 2

Publications & Pre-prints

Orhan, A.E. (2019). Robustness properties of Facebook’s ResNeXt WSL models. https://arxiv.org/ abs/1907.07640

Orhan, A.E., & Lake, B.M. (2019). Improving the robustness of ImageNet classifiers using elements of human visual cognition. https://arxiv.org/abs/1906.08416 (pre-print under review)

Orhan, A.E., & Pitkow, X. (2019). Improved memory in recurrent neural networks with sequential non-normal dynamics. https://arxiv.org/abs/1905.13715 (pre-print under review)

Orhan, A.E., & Ma, W.J. (2019) A diverse range of factors affect the nature of neural representations underlying short-term memory. Nature Neuroscience, 22, 275–283. [link]

Orhan, A.E., & Pitkow, X. (2018). Degeneracy, trainability, and generalization in deep neural networks. NeurIPS 2018 Workshop on Integration of Deep Learning Theories. [link]

Orhan, A.E. (2018). A simple cache model for image recognition. Advances in Neural Information Pro- cessing Systems 32 (NeurIPS 2018), https://arxiv.org/abs/1805.08709

Orhan, A.E., & Pitkow, X. (2018). Skip connections eliminate singularities. International Conference on Learning Representations (ICLR 2018), https://arxiv.org/abs/1701.09175

Orhan, A.E., & Ma, W.J. (2017). Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback. Nature Communications, 8, 138. [link]

Orhan, A.E., & Ma, W.J. (2015). Neural population coding of multiple stimuli. Journal of Neuroscience, 35(9), 3825-41. [link]

Orhan, A.E., & Jacobs, R.A. (2014). Are performance limitations in visual short-term memory tasks due to capacity limitations or model mismatch? https://arxiv.org/abs/1407.0644

Orhan, A.E., & Jacobs, R.A. (2014). Toward ecologically realistic theories in visual short-term memory research. Attention, Perception, & Psychophysics, 76, 2158-70. [link]

Orhan*, A.E., Sims*, C.R., Jacobs, R.A., & Knill, D.C. (2014). The adaptive nature of visual working memory. Current Directions in Psychological Science, 23(3), 164-70. (*equal contribution) [link]

Orhan, A.E., & Jacobs, R.A. (2013). A probabilistic clustering theory of the organization of visual short-term memory. Psychological Review, 120(2), 297-328. [link]

Orhan, A.E., & Jacobs, R.A. (2011). Probabilistic modeling of dependencies among visual short-term memory representations. Advances in Neural Information Processing Systems 24 (NIPS 2011). [link]

Orhan, A.E., & Jacobs, R.A. (2011). A nonparametric Bayesian model of visual short-term memory. Proceedings of the 33rd Annual Conference of the Cognitive Science Society (CogSci 2011). [link]

Orhan, A.E., Michel, M.M., & Jacobs, R.A. (2010). Visual learning with reliable and unreliable features. Journal of Vision, 10(2):2, 1-15. [link]

A. Emin Orhan 3

Conference Presentations

Orhan, A.E., & Pitkow, X. (2018). Degeneracy, trainability, and generalization in deep neural networks. Poster presented at the NeurIPS 2018 Workshop on Integration of Deep Learning Theories.

Orhan, A.E. (2018). A simple cache model for image recognition. Poster presented at NeurIPS 2018.

Orhan, A.E., & Pitkow, X. (2018). Skip connections eliminate singularities. Poster presented at ICLR 2018.

Orhan, A.E., & Ma, W.J. (2017). When do neural networks learn sequential solutions in short-term memory tasks? Talk presented at CCN 2017 (video: https://www.youtube.com/watch?v=KcpHCHGXBLE).

Orhan, A.E., & Ma, W.J. (2016). The inevitability of probability: probabilistic inference in generic neural networks trained with non-probabilistic feedback. Poster presented at COSYNE 2016.

Orhan, A.E. (2015). Connectivity shapes the dynamics of visual object recognition. Poster presented at COSYNE 2015.

Orhan, A.E., & Ma, W.J. (2014). Neural population coding of multiple stimuli. Poster presented at COSYNE 2014.

Orhan, A.E., & Jacobs, R.A. (2011). Probabilistic modeling of dependencies among visual short-term memory representations. Poster presented at NIPS 2011.

Orhan, A.E., & Jacobs, R.A. (2011). A nonparametric Bayesian model of visual short-term memory. Talk presented at CogSci 2011.

Invited Talks

New York University, Center for Data Science lunch seminar, May 8, 2019.

Columbia University, Neurotheory Center neural networks journal club, April 23, 2019.

New York University, Xiao-Jing Wang’s lab meeting, March 15, 2019.

Rice University, ML lunch seminar, February 21, 2018.

University of Houston, Networks seminar, September 23, 2016.

Johns Hopkins University, Vision seminar, March 14, 2014.

A. Emin Orhan 4

Academic/Research Experience

Participated in Machine Learning Summer School Cambridge, UK, September 2009.

Research Assistantship in Cognitive Science, Bogazici University Istanbul, Turkey, September 2005 - August 2008.

Participated in FIAS Summer School in Theoretical Neuroscience & Complex Systems Frankfurt, Germany, August 2007.

Teaching Experience

Teaching Assistant for BCS 111: Foundations of Cognitive Science University of Rochester, Fall 2009, Spring 2011, Spring 2012.

Reviewing Experience

Ad-hoc reviewer for: eLife, Psychonomic Bulletin & Review, CogSci, NeurIPS, ICLR

Awards/Grants

Moore-Sloan Data Science Fellowship, January 2019-current.

Travel grant and free registration for COSYNE 2016, February 2016.

Travel grant for NIPS 2011 Neural Information Processing Systems Foundation, December 2011.

Robert J. Glushko and Pamela Samuelson Foundation student travel grant Cognitive Science Society, July 2011.

Programming Skills

I do most of my day-to-day programming in Python. I use both TensorFlow and PyTorch for running neural network simulations. I am comfortable with shell scripting and have extensive experience in running large-scale simulations on HPC clusters.

References

Assist. Prof. Brenden M. Lake Department of Psychology & Center for Data Science, New York University Address: 60 5th Ave., Room 610, New York, NY 10011 Phone: +1 (212) 998-3059 Email: [email protected]

A. Emin Orhan 5

Assist. Prof. Xaq Pitkow Department of Neuroscience, Baylor College of Medicine Department of Electrical & Computer Engineering, Rice University Address: Baylor College of Medicine, One Baylor Plaza, rm T115, Houston TX 77030 Phone: +1 (978) 460-5144 Email: [email protected]

Assoc. Prof. Wei Ji Ma Center for Neural Science & Department of Psychology, New York University Address: Meyer Hall, 4 Washington Place, Room 730, New York, NY 10003 Phone: +1 (212) 992-6530 (office) Email: [email protected]

Prof. Robert A. Jacobs Department of Brain & Cognitive Sciences, University of Rochester Address: Meliora Hall 306, University of Rochester, Rochester, NY 14627-0268 Phone: +1 (585) 275-0753 (office) Email: [email protected]

Last updated: October 31, 2019