Human-in-the-Loop Interpretability Prior
Authors: Isaac Lage, Andrew Ross, Samuel J. Gershman, Been Kim, Finale Doshi-Velez
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We develop an algorithm that minimizes the number of user studies to find models that are both predictive and interpretable and demonstrate our approach on several data sets. Our human subjects results show trends towards different proxy notions of interpretability on different datasets. and 5 Experimental Setup, 6 Experimental Results. |
| Researcher Affiliation | Collaboration | Isaac Lage Department of Computer Science Harvard University isaaclage@g.harvard.edu Andrew Slavin Ross Department of Computer Science Harvard University andrew_ross@g.harvard.edu Been Kim Google Brain beenkim@google.com Samuel J. Gershman Department of Psychology Harvard University gershman@fas.harvard.edu Finale Doshi-Velez Department of Computer Science Harvard University finale@seas.harvard.edu |
| Pseudocode | No | The paper contains Figure 1 which is a high-level overview flowchart, but no structured pseudocode or algorithm blocks are present. |
| Open Source Code | No | The paper does not contain any statement about releasing source code for the methodology or provide a link to a code repository. |
| Open Datasets | Yes | We test our approach on a synthetic dataset as well as the mushroom, census income, and covertype datasets from the UCI database [7]. and [7] Dua Dheeru and EfiKarra Taniskidou. UCI machine learning repository, 2017. |
| Dataset Splits | Yes | Synthetic... We use an 80%-20% train-validate split., Mushroom... We use an 80%-20% train-validate split., Census... We use their 60%-40% train-validate split., Covertype... We use a 75%-25% train-validate split. |
| Hardware Specification | No | The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments. |
| Software Dependencies | No | The paper mentions the use of general machine learning models like 'decision trees' and 'neural networks' but does not provide specific software names with version numbers for libraries, frameworks, or solvers used for reproducibility. |
| Experiment Setup | No | Details of our model training procedure (that is, identifying models with high predictive accuracy) are in Appendix B. The covertype dataset, because it is modeled by a neural network, also needs a strategy for producing local explanations; we describe our parameter choices as well as provide a detailed sensitivity analysis to these choices in Appendix C. (This indicates setup details are in appendix, not main text). |