Optimal Decision Tree with Noisy Outcomes

Authors: Su Jia, viswanath nagarajan, Fatemeh Navidi, R Ravi

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate the performance of our algorithms on two natural applications with noise: toxic chemical identification and active learning of linear classifiers. Despite our theoretical logarithmic approximation guarantees, our methods give solutions with cost very close to the information theoretic minimum, demonstrating the effectiveness of our methods.
Researcher Affiliation Academia Su Jia Carnegie Mellon University sjia1@andrew.cmu.edu Fatemeh Navidi University of Michigan navidi@umich.edu Viswanath Nagarajan University of Michigan viswa@umich.edu R. Ravi Carnegie Mellon University ravi@andrew.cmu.edu
Pseudocode Yes Algorithm 1 ODTNr and Algorithm 2 Algorithm for ASR instance J are presented as structured algorithm blocks.
Open Source Code Yes The implementations of the adaptive and non-adaptive algorithms are available online.7 https://github.com/FatemehNavidi/ODTN ; https://github.com/sjia1/ODT-with-noisy-outcomes
Open Datasets Yes We considered a data set called WISER6, which includes 400+ chemicals (hypothesis) and 78 binary tests. Every chemical has either positive, negative or unknown result on each test. https://wiser.nlm.nih.gov
Dataset Splits No The paper describes different stopping criteria ('unique stopping', 'neighborhood stopping', 'clique stopping') and mentions using different versions of datasets (original vs. modified WISER), but it does not specify explicit training/validation/test dataset splits (e.g., percentages or counts) needed to reproduce the data partitioning.
Hardware Specification No The paper states 'We implemented our algorithms, and performed experiments', but does not provide any specific details about the hardware used (e.g., CPU, GPU, memory, or cloud instances).
Software Dependencies No The paper states 'The implementations... are available online' but does not specify any software dependencies or their version numbers (e.g., programming languages, libraries, or frameworks with versions).
Experiment Setup No The paper describes the datasets used and the algorithms implemented, and discusses different stopping criteria, but it does not provide specific experimental setup details such as hyperparameters (learning rates, batch sizes, epochs, optimizers, etc.) or other configuration settings.