Associative Long Short-Term Memory

Authors: Ivo Danihelka, Greg Wayne, Benigno Uria, Nal Kalchbrenner, Alex Graves

ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments in Section 6 show the benefits of the memory system for learning speed and accuracy.
Researcher Affiliation Industry Ivo Danihelka DANIHELKA@GOOGLE.COM Greg Wayne GREGWAYNE@GOOGLE.COM Benigno Uria BURIA@GOOGLE.COM Nal Kalchbrenner NALK@GOOGLE.COM Alex Graves GRAVESA@GOOGLE.COM Google Deep Mind
Pseudocode No No structured pseudocode or algorithm blocks are present. The paper describes methods using mathematical equations and prose.
Open Source Code No No statement about making source code publicly available or links to a code repository.
Open Datasets Yes We take a sequence of Image Net images (Russakovsky et al., 2015)...
Dataset Splits No For experiments with synthetic data, we generate new data for each training minibatch, obviating the need for a separate test data set.
Hardware Specification No No specific hardware details (e.g., CPU/GPU models, memory) are provided for the experimental setup.
Software Dependencies No The paper mentions "Adam optimisation algorithm" but does not provide specific software dependencies with version numbers for replication.
Experiment Setup Yes All experiments used the Adam optimisation algorithm (Kingma & Ba, 2014) with no gradient clipping. ... Minibatches of size 2 were used in all tasks beside the Wikipedia task below, where the minibatch size was 10.