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 Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
argmax centroid
Authors: Chengyue Gong, Mao Ye, Qiang Liu
NeurIPS 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the applicability and effectiveness of our method on a variety of real-world multitask learning applications, including few-shot image classification, personalized dialogue systems and multi-target domain adaptation. |
| Researcher Affiliation | Academia | Chengyue Gong Mao Ye Qiang Liu Computer Science Department, The University of Texas at Austin EMAIL |
| Pseudocode | Yes | Algorithm 1 Main Algorithm: Argmax Centroids for Approximating |
| Open Source Code | No | The paper does not provide any explicit statement about open-sourcing code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | Standard benchmarks of few-shot classification are chosen for experiments. We evaluate all the baselines and our algorithms on two subsets of Image Net, Mini-Image Net and Tiered Image Net (Sun et al., 2019). |
| Dataset Splits | Yes | Mini-Image Net contains 64 classes for training, 16 for validation and 20 for test. |
| Hardware Specification | No | The paper does not explicitly describe any specific hardware components (e.g., GPU/CPU models, memory specifications) used for running the experiments. |
| Software Dependencies | No | The paper mentions software components like 'BERT-base' and 'Adam' optimizer but does not provide specific version numbers for these or any other software dependencies needed for replication. |
| Experiment Setup | Yes | In all experiments, we set the replacement controller = 1.2 and = 0.5 for Algorithm 1. In the experiments, for few-shot learning based on SIB and IFSL, we set n = 16. For meta training, we use Adam (Kingma & Ba, 2014) with learning rate 10 3, 10 2 for inner and outer loop training, respectively. During the evaluation, for all the models, we used beam search with beam size 4 and length penalty 1.2. |