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 Shift Toward Open and Reproducible AI Research," Under Review, 2026. doi: 10.48550/arXiv.2606.16974..
Learning with Abandonment
Authors: Sven Schmit, Ramesh Johari
ICML 2018 | Venue PDF | LLM Run Details | Input Tokens: 15,272 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,556 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In Section 3.3, titled Simulations, the paper presents 'Cumulative regret plots' and states 'We observe that KL-UCB indeed performs better than the standard UCB algorithm.' It also states 'Code to replicate the simulations is available at https://github.com/schmit/learning-abandonment.' |
| Researcher Affiliation | Academia | Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA, USA 2Management Science & Engineering, Stanford University, Stanford, CA, USA. |
| Pseudocode | No | The paper describes algorithms and strategies in text but does not include any formal pseudocode blocks or clearly labeled algorithm sections. |
| Open Source Code | Yes | Code to replicate the simulations is available at https://github.com/schmit/learning-abandonment. |
| Open Datasets | No | The paper uses a simulated setting where 'the threshold distribution (unknown to the learning algorithm) is uniform on [0, 1]'. There is no traditional dataset provided with a URL, DOI, or repository, as the data is generated within the simulation based on this specified distribution. |
| Dataset Splits | No | The paper describes a simulation setup ('n = 2000 time steps', '50 repetitions') but does not specify traditional training, validation, or test dataset splits, as it operates within a sequential learning simulation environment rather than on a static dataset. |
| Hardware Specification | No | The paper does not specify any hardware used for running the simulations (e.g., CPU, GPU models, or cloud computing instances). |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | For the discretized policies, we set K βΌ 2(n/ log n)^1/4 = 10. The explore-exploit strategy ο¬rst observes 20 + 2 β n = 110 samples to estimate F, before committing to a ο¬xed strategy. |