Unsupervised learning of an efficient short-term memory network
Authors: Pietro Vertechi, Wieland Brendel, Christian K. Machens
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We simulated a firing rate network of ten neurons that learn to remember a one-dimensional, temporally uncorrelated white noise stimulus (Fig. 2). |
| Researcher Affiliation | Academia | Pietro Vertechi Wieland Brendel Christian K. Machens Champalimaud Neuroscience Programme Champalimaud Centre for the Unknown Lisbon, Portugal first.last@neuro.fchampalimaud.org current address: Centre for Integrative Neuroscience, University of T ubingen, Germany |
| Pseudocode | No | The paper describes the learning rules and dynamics using mathematical equations and prose but does not include any clearly labeled pseudocode blocks or algorithms. |
| Open Source Code | No | The paper does not provide any statement about releasing source code or a link to a code repository. |
| Open Datasets | No | The paper states that it 'simulated a firing rate network of ten neurons that learn to remember a one-dimensional, temporally uncorrelated white noise stimulus'. This indicates a generated stimulus rather than a publicly available dataset with concrete access information. |
| Dataset Splits | No | The paper describes a simulation with a generated white noise stimulus but does not specify any training, validation, or test dataset splits. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for simulations, such as CPU/GPU models or memory specifications. |
| Software Dependencies | No | The paper does not list any specific software dependencies with version numbers, such as programming languages, libraries, or frameworks used for the simulations. |
| Experiment Setup | Yes | We initialized all feedforward weights to one, whereas the matrices Ωf and Ωd were initialised by drawing numbers from centered Gaussian distributions with variance 1 and 0.2 respectively. All matrices were then divided by N 2 = 100. Firing rates were constrained to be positive. |