Feedback Detection for Live Predictors

Authors: Stefan Wager, Nick Chamandy, Omkar Muralidharan, Amir Najmi

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct a pilot study for our proposed methodology using a predictive system currently deployed as a part of a search engine. [...] Finally, in Section 5 we conduct a pilot study based on a predictive model currently deployed as a part of a search engine.
Researcher Affiliation Collaboration Stefan Wager, Nick Chamandy, Omkar Muralidharan, and Amir Najmi swager@stanford.edu, {chamandy, omuralidharan, amir}@google.com Stanford University and Google, Inc.
Pseudocode No The paper describes the steps of the method under 'Our Method in Practice' but does not provide a formal pseudocode block or algorithm listing.
Open Source Code No The paper does not provide any links to open-source code or explicitly state that the code for their methodology is available.
Open Datasets No The paper mentions using 'historical data collected from log files' and an 'internal' dataset for a 'predictive system currently deployed as a part of a search engine'. No public access information, links, or citations for a publicly available dataset are provided.
Dataset Splits Yes Our dataset had on the order of 100,000 data points, half of which were used for fitting the model itself and half of which were used for feedback simulation.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments.
Software Dependencies No The paper mentions 'standard R libraries' but does not specify any software names with version numbers.
Experiment Setup Yes We generated data for 5 simulated time periods, adding noise with σ = 0.1 at each step, and fit feedback using a spline basis discussed in Appendix B. The true feedback curve was obtained by fitting a spline regression to the additive feedback model by looking at the unobservable ˆy(t+1) i [?]; we used a df = 5 natural spline with knots evenly spread out on [ 9, 3] in log-odds space plus a jump at 0. [...] The error bars for estimated feedback were obtained using a non-parametric bootstrap [11] for which we resampled pairs of (current, next) predictions.