Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Approximating a Distribution Using Weight Queries
Authors: Nadav Barak, Sivan Sabato
ICML 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To supplement our theoretical analysis, we also imple-ment the proposed algorithm and report several experiments, which demonstrate its advantage over several natural baselines. A python implementation of the proposed algorithm and of all the experiments can be found at https://github.com/Nadav-Barak/AWP. |
| Researcher Affiliation | Academia | 1Department of Computer Science, Ben Gurion University, Israel. Correspondence to: Nadav Barak <EMAIL>, Sivan Sabato <EMAIL>. |
| Pseudocode | Yes | Algorithm 1 AWP: Approximated weighting of a data set via a low-discrepancy pruning |
| Open Source Code | Yes | A python implementation of the proposed algorithm and of all the experiments can be found at https://github.com/Nadav-Barak/AWP. |
| Open Datasets | Yes | In the first set of experiments, the input data set was Adult (Dua and Graff, 2017)... First, we tested the MNIST (Le Cun and Cortes, 2010) training set... The input data set was Caltech256 (Griffin et al., 2007)... We used two target data sets: (1) The Office data set (Saenko et al., 2010)... (2) The Bing data set (Alessandro Bergamo, 2010a;b)... |
| Dataset Splits | No | The paper mentions datasets like 'MNIST training set' but does not specify the explicit breakdown of these datasets into training, validation, or test sets by percentages, sample counts, or specific splitting methodologies. |
| Hardware Specification | No | The paper states, 'The implementation can be easily run on a standard personal computer, with the longest runs taking a few hours.' This is too vague and does not provide specific hardware details (e.g., CPU, GPU models, or memory). |
| Software Dependencies | No | The paper only mentions 'A python implementation' but does not specify the version of Python or any specific library names with their version numbers that are necessary for replication. |
| Experiment Setup | Yes | We ran several types of experiments, all with inputs δ = 0.05 and β = 4. |