no code implementations • 11 Jul 2023 • Kyle Luther, H. Sebastian Seung
A recent trend is to train a convolutional net to denoise images, and use this net as a prior when solving the inverse problem.
no code implementations • 16 Jun 2023 • Kyle Luther, Sebastian Seung
We compare our method to using a U-Net to directly reconstruct the unstretched tilt views and show that this simple stretching procedure leads to significantly better reconstructions.
no code implementations • 6 Jun 2022 • Kyle Luther, H. Sebastian Seung
We regard such dependence of unsupervised learning on prior knowledge implicit in network architecture as biologically plausible, and analogous to the dependence of brain architecture on evolutionary history.
no code implementations • 15 Apr 2022 • Kyle Luther, H. Sebastian Seung
We show empirically with the MNIST dataset that sparse codes can be very sensitive to image distortions, a behavior that may hinder invariant object recognition.
no code implementations • 15 Apr 2022 • Kyle Luther, H. Sebastian Seung
Recent works have derived neural networks with online correlation-based learning rules to perform \textit{kernel similarity matching}.
no code implementations • 29 Sep 2021 • Kyle Luther, Sebastian Seung
Minimizing this upper bound leads to a minimax optimization, which can be solved via stochastic gradient descent-ascent.
no code implementations • 21 Sep 2019 • Kisuk Lee, Ran Lu, Kyle Luther, H. Sebastian Seung
We show dense voxel embeddings learned via deep metric learning can be employed to produce a highly accurate segmentation of neurons from 3D electron microscopy images.
no code implementations • 13 Feb 2019 • Kyle Luther, H. Sebastian Seung
Before training a neural net, a classic rule of thumb is to randomly initialize the weights so the variance of activations is preserved across layers.
no code implementations • 31 Jan 2019 • Kyle Luther, H. Sebastian Seung
Both empirically and theoretically, it is unclear whether or when deep metric learning is superior to the more conventional approach of directly predicting an affinity graph with a convolutional net.
Electron Microscopy Image Segmentation Image Segmentation +3