Neurons Equipped with Intrinsic Plasticity Learn Stimulus Intensity Statistics
Authors: Travis Monk, Cristina Savin, Jörg Lücke
NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Numerical experiments verify our analytical derivations and show robust behavior for artificial and natural stimuli. |
| Researcher Affiliation | Academia | Travis Monk Cluster of Excellence Hearing4all University of Oldenburg 26129 Oldenburg, Germany travis.monk@uol.de Cristina Savin IST Austria 3400 Klosterneuburg Austria csavin@ist.ac.at J org L ucke Cluster of Excellence Hearing4all University of Oldenburg 26129 Oldenburg, Germany joerg.luecke@uol.de |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain an explicit statement about releasing source code or provide a link to a code repository for the described methodology. |
| Open Datasets | Yes | To test the neural network on more realistic data, we followed a number of related studies [8 12] and used the MNIST as a standard dataset containing different stimulus classes. |
| Dataset Splits | No | The paper mentions using a 'modified MNIST training set' for initialization and '0.5% (thin lines) and 5% (thick lines) of labels in the training set' for calculating classification success rates, but it does not specify explicit training, validation, and test dataset splits with percentages or sample counts for the overall experimental data partitioning. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU or CPU models, memory, or cloud resources used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | Yes | For these plots, ϵW = ϵλ = .005, D = 100 (for a 10x10 pixel grid), C = 4, initialized weights were uniformly-distributed between .01 and .06, and initialized intrinsic parameters were uniformly-distributed between 10 and 20. (...) The λc were initialized to be the mean intensity of all digit classes as calculated from our modified MNIST training set. Each Wcd was initialized as Wcd Pois(Wcd; µd) + 1, where µd is the mean of each pixel over all classes and is calculated from our modified MNIST training set. |