Methods for modeling brain structure and organization
The lab is actively developing computer vision, ML, and deep learning methods for analyzing large-scale neuroimaging datasets and building models of neuroanatomy and brain organization. Some examples include: mapping and discovery of brain areas, learning cytoarchitecture + layers in cortical and retinal tissues, and modeling high-resolution spatial maps of whole-brain connectivity.
- A.H. Balwani, E.L. Dyer, Modeling variability in brain architecture with deep feature learning, Proc. of the IEEE Asilomar Conference on Signals, Systems, and Computers, 2019 (Paper)
- E.L. Dyer et al., Quantifying mesoscale neuroanatomy using X-ray microtomography, Web, Paper) , 2017 (
- T.J. LaGrow, M. Moore, J.A. Prasad, A. Webber , M.A. Davenport, E.L. Dyer, Cytoarchitecture and Layer Estimation in High-Resolution Neuroanatomical Images, bioarXiv, July 2018 (Preprint)
- D. Rolnick, E.L. Dyer, Generative models and abstractions for large-scale neuroanatomy datasets, Current Opinion in Neurobiology, February 2018. (Paper, Current Opinion)
- M. Dabagia, E.L. Dyer, Barycenters in the brain: An optimal transport approach for modeling connectivity, Optimal Transport in Machine Learning Workshop, NeurIPS, 2019.
Modeling and comparing high-dimensional neural datasets with distribution alignment
In another line of research, the lab is developing methods for transfer learning and distribution alignment of datasets across domains. We are actively pursuing the application of these techniques for comparing and aligning neural datasets.
- J. Lee, M. Dabagia, E.L. Dyer*, C. Rozell*, Hierarchical Optimal Transport for Multimodal Distribution Alignment, Neural Information Processing Systems (NeurIPS), Dec 2019. (Preprint, Python Code)
- E.L. Dyer, M. Azar, H.L. Fernandes, M. Perich, L.E. Miller, and K.P. Körding: A cryptography-based approach to brain decoding, Nature Biomedical Engineering, 2017. (Web, Paper)
- M. Dabagia, K.P. Kording, E.L. Dyer,
- C-H. Lin, M. Azabou, E.L. Dyer, Making transport more robust and interpretable by moving data through a small number of anchor points, Dec 2020 (Preprint)
Low-dimensional signal models
Unions of subspaces (UoS) are a generalization of single subspace models that approximate data points as living on multiple subspaces, rather than assuming a global low-dimensional model (as in PCA). Modeling data with mixtures of subspaces provides a more compact and simple representation of the data, and thus can lead to better partitioning (clustering) of the data and help in compression and denoising.
- E.L. Dyer, A.C. Sankaranarayanan, and R.G. Baraniuk, Greedy feature selection for subspace clustering, The Journal of Machine Learning Research 14 (1), 2487-2517, September, 2013. (Paper)
- E.L. Dyer, T.A. Goldstein, R. Patel, K.P. Körding, and R.G. Baraniuk, Sparse self-expressive decompositions for dimensionality reduction and clustering (Paper)
- R.J. Patel, T.A. Goldstein, E.L. Dyer, A. Mirhoseini, and R.G. Baraniuk, Deterministic column sampling for low rank approximation: Nystrom vs. Incomplete Cholesky Decomposition, SIAM Data Mining (SDM) Conference, May 2016. (Paper, Code)
Large-scale and non-convex optimization
Optimization problems are ubiquitous in machine learning and neuroscience. The lab works on a few different topics in the areas of non-convex optimization and distributed machine learning.
- C-H Lin, JD Miano, EL Dyer, Bayesian optimization for modular black-box systems with switching costs, preprint coming soon, May 2020
- A. Mirhoseini, E.L. Dyer, E. Songhori, R.G. Baraniuk, and F. Koushanfar, RankMap: A platform-aware framework for distributed learning from dense datasets, IEEE Trans. on Neural Networks and Learning Systems, 2017. (Paper, Code)
- M Gheshlaghi Azar, E.L. Dyer, Konrad Kording, Convex Relaxation Regression (CoRR): Black-box optimization of a smooth function by learning its convex envelope, Proc. of the Conference on Uncertainity in Artificial Intelligence, 2016. (Paper)