Planning with Biological Neurons and Synapses
Authors: Francesco d'Amore, Daniel Mitropolsky, Pierluigi Crescenzi, Emanuele Natale, Christos H. Papadimitriou21-28
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The purpose of this work is to establish empirically that reasonably large programs in the Assembly Calculus can execute correctly and reliably; and that rather realistic if idealized higher cognitive functions, such as planning in the blocks world, can be implemented successfully by such programs. |
| Researcher Affiliation | Academia | 1 Universit e Cˆote d Azur, Inria, CNRS, I3S, Sophia Antipolis, France 2 Department of Computer Science, Columbia University, New York, USA 3 Gran Sasso Science Institute, L Aquila, Italia |
| Pseudocode | Yes | Algorithm 1: PARSER (S) input: a stack S of blocks b1, b2, . . . , bs. |
| Open Source Code | Yes | The Julia source code can be found at (d Amore et al. 2021a). |
| Open Datasets | No | The paper states, 'We make use of the Java generator for BW configurations available at Koeman (2020), based on Slaney and Thi ebaux (2001).' This refers to a tool that generates configurations, but it does not provide concrete access to a pre-existing, publicly available dataset used for the experiments. |
| Dataset Splits | No | The paper mentions running experiments on 'over 100 blocks-world configurations' but does not specify any training, validation, or test dataset splits or how these configurations were partitioned for evaluation. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as CPU or GPU models, or memory specifications. |
| Software Dependencies | No | The paper states, 'A software system for programming in the AC, as well as implementations of the algorithms described in this paper, have been written in Julia (Bezanson et al. 2017).' While Julia is mentioned with a reference, no specific libraries, frameworks, or solvers with version numbers are listed as software dependencies. |
| Experiment Setup | Yes | We have used various settings of the parameters n, k, p, β a particularly good set of parameters is n = 106, k = 50, p = 0.1, β = 0.1. |