Nicholas Turner will present his generals exam on Monday, May 16, 2016 at 2pm in CS 402. The members of his committee are Sebastian Seung (adviser), Tom Funkhouser, and Jianxiong Xiao Everyone is invited to attend his talk, and those faculty wishing to remain for the oral exam following are welcome to do so. His abstract and reading list follow below. *Abstract:* Connectomics studies neural circuits by extracting their structural properties from electron microscopy (EM) images. Current state-of-the-art approaches use convolutional neural networks as a step towards segmenting aligned stacks of EM images. One application of these methods attempts to study the mouse retina. However, current practices have featured poor performance in describing the geometry of ganglion cell somata, which are difficult to segment as they can feature large image staining gaps. I will present work addressing two related challenges. First, I will demonstrate how we can preprocess the image stack to improve the segmentation accuracy of current approaches. These improvements allow us to quantify basic properties of ganglion cell somata. Second, I will present preliminary work motivated by these results to provide segmentation across multiple layers of tissue by means of a single multiscale model. *Reading List:* Bishop, Christopher M. "Pattern Recognition." *Machine Learning* (2006). Ian Goodfellow, Yoshua Bengio, and Aaron Courville. "Deep Learning." Book in preparation for MIT Press, 2016. J. S. Kim, M. J. Greene, A. Zlateski, K. Lee, M. Richardson, S. C. Turaga, M. Purcaro, M. Balkam, A. Robinson, B. F. Behabadi, M. Campos, W. Denk, H. S. Seung, and EyeWirers. Space-time wiring specificity supports direction selectivity in the retina. http://www.ncbi.nlm.nih.gov/pubmed/24805243 *Nature**509*, 331-6 (2014). Sanes, Joshua R., and Richard H. Masland. "The types of retinal ganglion cells: current status and implications for neuronal classification." *Annual review of neuroscience* 38 (2015): 221-246. Ciresan, Dan, et al. "Deep neural networks segment neuronal membranes in electron microscopy images." *Advances in neural information processing systems*. 2012. S. C. Turaga, K. L. Briggman, M. Helmstaedter, W. Denk, and H. S. Seung. *Maximin affinity learning of image segmentation*. CoRR. abs/0911.5372 (2009). S. C. Turaga, J. F. Murray, V. Jain, F. Roth, M. Helmstaedter, K. Briggman, W. Denk, and H. S. Seung. Convolutional networks can learn to generate affinity graphs for image segmentation. http://www.ncbi.nlm.nih.gov/pubmed/19922289*Neural Comput.* *22*, 511-38 (2010). Glorot, Xavier, and Yoshua Bengio. “Understanding the difficulty of training deep feedforward neural networks.” http://jmlr.csail.mit.edu/proceedings/papers/v9/glorot10a/glorot10a.pdf *Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS’10). Society for Artificial Intelligence and Statistics*. 2010. Bengio, Yoshua. "Learning deep architectures for AI." *Foundations and trends® in Machine Learning* 2.1 (2009): 1-127. I. Arganda-Carreras, S.C. Turaga, D.R. Berger, D. Cireşan, A. Giusti, L.M. Gambardella, J. Schimdhuber, D. Laptev, S. Dwivedi, J.M. Buhmann, T. Liu, M. Seyedhosseini, T. Tasdizen, L. Kamentsky, R. Burget, V. Uher, X. Tan, C. Sun, T.D. Pham, E. Bas, M.G. Uzunbas, A. Cardona, J. Schindelin, H.S. Seung. Crowdsourcing the creation of image segmentation algorithms for connectomics http://journal.frontiersin.org/article/10.3389/fnana.2015.00142/abstract. Front Neuroanat *142*, (2015).