Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
175 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Signal Processing in the Retina: Interpretable Graph Classifier to Predict Ganglion Cell Responses (2401.01813v1)

Published 3 Jan 2024 in cs.LG, eess.IV, q-bio.NC, and q-bio.QM

Abstract: It is a popular hypothesis in neuroscience that ganglion cells in the retina are activated by selectively detecting visual features in an observed scene. While ganglion cell firings can be predicted via data-trained deep neural nets, the networks remain indecipherable, thus providing little understanding of the cells' underlying operations. To extract knowledge from the cell firings, in this paper we learn an interpretable graph-based classifier from data to predict the firings of ganglion cells in response to visual stimuli. Specifically, we learn a positive semi-definite (PSD) metric matrix $\mathbf{M} \succeq 0$ that defines Mahalanobis distances between graph nodes (visual events) endowed with pre-computed feature vectors; the computed inter-node distances lead to edge weights and a combinatorial graph that is amenable to binary classification. Mathematically, we define the objective of metric matrix $\mathbf{M}$ optimization using a graph adaptation of large margin nearest neighbor (LMNN), which is rewritten as a semi-definite programming (SDP) problem. We solve it efficiently via a fast approximation called Gershgorin disc perfect alignment (GDPA) linearization. The learned metric matrix $\mathbf{M}$ provides interpretability: important features are identified along $\mathbf{M}$'s diagonal, and their mutual relationships are inferred from off-diagonal terms. Our fast metric learning framework can be applied to other biological systems with pre-chosen features that require interpretation.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. L. A. Remington, Clinical Anatomy of the Visual System (2nd ed.). Elsevier, 2005.
  2. T. Gollisch and M. Meister, “Eye smarter than scientists believed: neural computations in circuits of the retina,” Neuron, vol. 65, no. 2, pp. 150–164, 2010.
  3. L. McIntosh, N. Maheswaranathan, A. Nayebi, S. Ganguli, and S. Baccus, “Deep learning models of the retinal response to natural scenes,” Advances in Neural Information Processing Systems, vol. 29, 2016.
  4. A. Lozano, C. Soto-Sanchez, J. Garrigos, J. J. Martinez, J. M. Ferrández, and E. Fernandez, “A 3d convolutional neural network to model retinal ganglion cell’s responses to light patterns in mice,” International Journal of Neural Systems, vol. 28, no. 10, p. 1850043, 2018.
  5. H. Tanaka, A. Nayebi, N. Maheswaranathan, L. McIntosh, S. Baccus, and S. Ganguli, “From deep learning to mechanistic understanding in neuroscience: the structure of retinal prediction,” Advances in Neural Information Processing Systems, vol. 32, 2019.
  6. E. Batty, J. Merel, N. Brackbill, A. Heitman, A. Sher, A. Litke, E. Chichilnisky, and L. Paninski, “Multilayer recurrent network models of primate retinal ganglion cell responses,” in International Conference on Learning Representations, 2017.
  7. J. Pang and G. Cheung, “Graph Laplacian regularization for image denoising: Analysis in the continuous domain,” IEEE Transactions on Image Processing, vol. 26, no. 4, pp. 1770–1785, 2017.
  8. L. Yang and R. Jin, “Distance metric learning: A comprehensive survey,” Michigan State Universiy, vol. 2, no. 2, p. 4, 2006.
  9. K. Q. Weinberger and L. K. Saul, “Distance metric learning for large margin nearest neighbor classification.,” Journal of Machine Learning Research, vol. 10, no. 2, 2009.
  10. Z.-q. Luo, W.-k. Ma, A. M.-c. So, Y. Ye, and S. Zhang, “Semidefinite relaxation of quadratic optimization problems,” IEEE Signal Processing Magazine, vol. 27, no. 3, pp. 20–34, 2010.
  11. C. Yang, G. Cheung, and W. Hu, “Signed graph metric learning via Gershgorin disc perfect alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 10, pp. 7219–7234, 2022.
  12. D. Marshak, “Retinal ganglion cells: Anatomy,” in Encyclopedia of Neuroscience, pp. 211–218, Oxford: Academic Press, 2009.
  13. G. Zeck, M. Bethge, and J. H. Macke, “Receptive fields without spike-triggering,” in Advances in Neural Information Processing Systems, vol. 20, Curran Associates Inc., 2007.
  14. J. B. Demb, K. Zaghloul, L. Haarsma, and P. Sterling, “Bipolar cells contribute to nonlinear spatial summation in the brisk-transient (y) ganglion cell in mammalian retina,” Journal of Neuroscience, vol. 21, no. 19, pp. 7447–7454, 2001.
  15. M. J. Berry, I. H. Brivanlou, T. A. Jordan, and M. Meister, “Anticipation of moving stimuli by the retina,” Nature, vol. 398, no. 6725, pp. 334–338, 1999.
  16. B. P. Ölveczky, S. A. Baccus, and M. Meister, “Segregation of object and background motion in the retina,” Nature, vol. 423, no. 6938, pp. 401–408, 2003.
  17. T. A. Münch, R. A. Da Silveira, S. Siegert, T. J. Viney, G. B. Awatramani, and B. Roska, “Approach sensitivity in the retina processed by a multifunctional neural circuit,” Nature Neuroscience, vol. 12, no. 10, pp. 1308–1316, 2009.
  18. E. Chichilnisky, “A simple white noise analysis of neuronal light responses,” Network: Computation in Neural Systems, vol. 12, no. 2, p. 199, 2001.
  19. A. Heitman, N. Brackbill, M. Greschner, A. Sher, A. M. Litke, and E. Chichilnisky, “Testing pseudo-linear models of responses to natural scenes in primate retina,” BioRxiv, p. 045336, 2016.
  20. P. Linardatos, V. Papastefanopoulos, and S. Kotsiantis, “Explainable AI: A review of machine learning interpretability methods,” Entropy, vol. 23, no. 1, p. 18, 2020.
  21. A. Ortega, P. Frossard, J. Kovacevic, J. M. F. Moura, and P. Vandergheynst, “Graph signal processing: Overview, challenges, and applications,” Proceedings of the IEEE, vol. 106, no. 5, pp. 808–828, 2018.
  22. G. Cheung, E. Magli, Y. Tanaka, and M. K. Ng, “Graph spectral image processing,” Proceedings of the IEEE, vol. 106, no. 5, pp. 907–930, 2018.
  23. W. Hu, X. Gao, G. Cheung, and Z. Guo, “Feature graph learning for 3d point cloud denoising,” IEEE Transactions on Signal Processing, vol. 68, pp. 2841–2856, 2020.
  24. X. Li, C. Yang, W. Tong, F. Shi, and G. Zhai, “Fast graph-based binary classifier learning via further relaxation of semi-definite relaxation,” in Proceedings of the 2022 5th International Conference on Image and Graphics Processing, pp. 89–95, 2022.
  25. O. Marre, G. Tkacik, D. Amodei, E. Schneidman, W. Bialek, and M. Berry, “Multi-electrode array recording from salamander retinal ganglion cells,” 2017. IST Austria, 10.15479/AT:ISTA:61.
  26. G. Tkacik, O. Marre, D. Amodei, E. Schneidman, W. Bialek, and M. J. Berry, “Searching for collective behavior in a large network of sensory neurons,” PLoS Computational Biology, vol. 10, no. 1, p. e1003408, 2014.
  27. D. G. Lowe, “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision, vol. 60, pp. 91–110, 2004.
  28. S. Jiang, Z. Song, O. Weinstein, and H. Zhang, “Faster dynamic matrix inverse for faster LPs,” arXiv preprint arXiv:2004.07470, 2020.
  29. Springer Science & Business Media, 2010.
  30. D. Cartwright and F. Harary, “Structural balance: a generalization of Heider’s theory,” in Psychological Review, vol. 63, no.5, pp. 277–293, 1956.
  31. A. V. Knyazev, “Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method,” SIAM Journal on Scientific Computing, vol. 23, no. 2, pp. 517–541, 2001.
  32. H. F. et al., “PyTorchVideo: A deep learning library for video understanding,” in Proceedings of the 29th ACM International Conference on Multimedia, 2021. https://pytorchvideo.org/.
  33. C. Feichtenhofer, H. Fan, J. Malik, and K. He, “Slowfast networks for video recognition,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6202–6211, 2019.
  34. I. Hadji and R. P. Wildes, “A spatiotemporal oriented energy network for dynamic texture recognition,” in Proceedings of the IEEE International Conference on Computer Vision, pp. 3066–3074, 2017.
  35. I. Hadji, “SOE-Net.” https://github.com/hadjisma/SOE-Net, 2020.
  36. B. Rister, M. A. Horowitz, and D. L. Rubin, “Volumetric image registration from invariant keypoints,” IEEE Transactions on Image Processing, vol. 26, no. 10, pp. 4900–4910, 2017.
  37. B. Rister, “Sift3d.” https://github.com/bbrister/SIFT3D, 2019.
  38. L. Deng, “The mnist database of handwritten digit images for machine learning research,” IEEE Signal Processing Magazine, vol. 29, no. 6, pp. 141–142, 2012.
  39. T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794, 2016.
  40. J. H. Friedman, “Greedy function approximation: a gradient boosting machine,” Annals of Statistics, pp. 1189–1232, 2001.
  41. John Wiley & Sons, 2013.
  42. T. Cover and P. Hart, “Nearest neighbor pattern classification,” IEEE Transactions on Information Theory, vol. 13, no. 1, pp. 21–27, 1967.
  43. J. Strum, “A matlab toolbox for optimization over symmetric cones,” Using SEDUMI 1.02 (ver. 1.05), 2001.
  44. J. Goetz, Z. F. Jessen, A. Jacobi, A. Mani, S. Cooler, D. Greer, S. Kadri, J. Segal, K. Shekhar, J. R. Sanes, et al., “Unified classification of mouse retinal ganglion cells using function, morphology, and gene expression,” Cell Reports, vol. 40, no. 2, 2022. http://rgctypes.org/.
  45. J. Shlens, G. D. Field, J. L. Gauthier, M. I. Grivich, D. Petrusca, A. Sher, A. M. Litke, and E. Chichilnisky, “The structure of multi-neuron firing patterns in primate retina,” Journal of Neuroscience, vol. 26, no. 32, pp. 8254–8266, 2006.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now