Do Not Have Enough Data? Deep Learning to the Rescue!

Authors: Ateret Anaby-Tavor, Boaz Carmeli, Esther Goldbraich, Amir Kantor, George Kour, Segev Shlomov, Naama Tepper, Naama Zwerdling7383-7390

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In a series of experiments, we show that LAMBADA improves classifiers performance on a variety of datasets.
Researcher Affiliation Collaboration 1IBM Research AI, 2University of Haifa, Israel, 3Technion Israel Institute of Technology
Pseudocode Yes We define the method in Algorithm 1 and elaborate on its steps in the following section.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes ATIS Flight reservations 17 4.2k (www.kaggle.com/siddhadev/atis-dataset-from-ms-cntk); TREC Open-domain questions 50 6k (https://cogcomp.seas.upenn.edu/Data/QA/QC/)
Dataset Splits Yes We randomly split each dataset into train, validation, and test sets (80%, 10%, 10%).
Hardware Specification No The paper does not specify the exact hardware (e.g., GPU models, CPU types) used for running the experiments.
Software Dependencies No The paper mentions software components like BERT, SVM, LSTM, and GloVe but does not provide specific version numbers for these or other software dependencies.
Experiment Setup No The paper describes model architectures and general settings (e.g., GloVe 100 dimensions) but lacks specific hyperparameter values (e.g., learning rate, batch size, number of epochs, optimizer settings) or detailed system-level training configurations to reproduce the experiment setup.