mwspratling
- https://nms.kcl.ac.uk/michael.spratling/
-
I've been researching neural networks since 1994, and have built models ranging from ones consisting of a few neurons to evaluate theories of cortical computation, to ones containing billions of parameters applied to problems in computer vision.
- Joined on
2023-02-10
Code to comprehensively evaluate the robustness of pytorch classifiers
Updated 2023-12-14 14:12:32 +00:00
Code to evaluate the adversarial defense provided by the Symmetric Divisive Normalisation activation function
Updated 2023-02-14 10:18:01 +00:00
Code for finding matching image patches using Divisive Input Modulation, and competing methods.
Updated 2023-02-14 08:47:31 +00:00
Code for simulating the response properties of V1 mismatch neurons and for testing the ability of predictive coding algorithms to scale.
Updated 2023-02-14 08:46:46 +00:00
Code for learning in a hierarchical PC/BC-DIM model. The model is applied to learning the bars problem, and to learning using natural images. In the latter case RFs similar to those found in cortical areas V1 and V2 are learnt.
Updated 2023-02-14 08:46:14 +00:00
Code for modelling reverse correlation experiments using the PC/BC-DIM model of V1.
Updated 2023-02-14 08:45:42 +00:00
Code for learning the squares problem using the divisive input modulation (DIM) algorithm.
Updated 2023-02-14 08:45:16 +00:00
Code for performing Hough voting with explaining away using the PC/BC-DIM algorithm.
Updated 2023-02-14 08:44:21 +00:00
Code for simulating a variety of behavioural and physiological phenomena related to gaze control with the PC/BC-DIM algorithm.
Updated 2023-02-14 08:43:49 +00:00
Code for performing object recognition with the PC/BC-DIM algorithm.
Updated 2023-02-14 08:43:23 +00:00
Code for performing Bayesian inference using the PC/BC-DIM algorithm.
Updated 2023-02-14 08:42:40 +00:00
Code for performing pattern recognition using a range of different sparse coding algorithms, including the PC/BC-DIM algorithm.
Updated 2023-02-14 08:42:22 +00:00