Large-Scale Optimistic Adaptive Submodularity

Authors: Victor Gabillon, Branislav Kveton, Zheng Wen, Brian Eriksson, S. Muthukrishnan

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

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we evaluate our solution on two problems, preference elicitation and face detection, and show that high-quality policies can be learned sample efficiently. ... Lin OASM is evaluated on two real-world problems. The first problem is learning of a policy for movie recommendation. The second problem is learning of an adaptive face detection policy.
Researcher Affiliation Collaboration Victor Gabillon INRIA Lille team Seque L Villeneuve d Ascq, France victor.gabillon@inria.fr Branislav Kveton Technicolor Labs Palo Alto, CA branislav.kveton@technicolor.com Zheng Wen Electrical Engineering Department Stanford University zhengwen@stanford.edu Brian Eriksson Technicolor Labs Palo Alto, CA brian.eriksson@technicolor.com S. Muthukrishnan Department of Computer Science Rutgers muthu@cs.rutgers.edu
Pseudocode Yes Algorithm 1 Lin OASM: Optimistic adaptive submodularity with a generalized linear model.
Open Source Code No The paper does not contain any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes We experiment with 6k users and 500 most rated movies from the Movie Lens dataset (Lam and Herlocker 2013). ... We experiment with 3k labeled images of faces from the GENKI-SZSL dataset (UCSD 2013).
Dataset Splits No The paper describes an episodic learning process but does not specify explicit train/validation/test splits of the datasets with percentages or sample counts for reproducibility.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU, CPU models) used for running the experiments.
Software Dependencies No The paper does not specify any software dependencies (e.g., libraries, frameworks) with version numbers.
Experiment Setup Yes In our experiments, we choose λ = 1 and ρk,t(δ) = 0.1 log(t K + k).