Efficient Partial Monitoring with Prior Information
Authors: Hastagiri P Vanchinathan, Gábor Bartók, Andreas Krause
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | As demonstrated with experiments on synthetic as well as real-world data, the algorithm outperforms previous approaches, even for very uninformed priors, with an order of magnitude smaller regret and lower running time. [...] First, we run extensive evaluations of BPM on various synthetic datasets and compare the performance against CBP [10] and Feed Exp3 [7]. |
| Researcher Affiliation | Academia | Hastagiri P Vanchinathan Dept. of Computer Science ETH Z urich, Switzerland hastagiri@inf.ethz.ch G abor Bart ok Dept. of Computer Science ETH Z urich, Switzerland bartok@inf.ethz.ch Andreas Krause Dept. of Computer Science ETH Z urich, Switzerland krausea@ethz.ch |
| Pseudocode | Yes | Pseudocode for the algorithm family is shown in Algorithm 1. [...] Algorithm 1 BPM input: L,H,p0,Σ0 initialization: Calculate signal matrices Si for t=1 to T do Use selection rule (cf., Sec. 3.2) to choose an action It Observe feedback Yt Update posterior: Σ 1 t =Σ 1 t 1+PIt and pt=Σt Σ 1 t 1pt 1+S It(SIt S It) 1Yt ; end for |
| Open Source Code | No | The paper does not include any explicit statement about making the source code available or provide a link to a code repository. |
| Open Datasets | Yes | We also provide results of BPM on a dataset that was collected by Singla and Krause [13] from real interactions with many users on the Amazon Mechanical Turk (AMT) [14] crowdsourcing platform. [...] The datasets used in the simulated experiments are identical to the ones used by Bart ok et al. [10] and thus allow us to benchmark against the current state of the art. |
| Dataset Splits | No | The paper describes simulating a stream of users ('a stream of 300000 potential users') for experiments but does not provide explicit training, validation, or test set percentages or sample counts for dataset splits. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies or version numbers (e.g., programming languages, libraries, frameworks) used for the implementation or execution of the experiments. |
| Experiment Setup | Yes | To use BPM-LEAST (see Section 3.2), we need to recover the current feasible actions. We do so by sampling multiple (10000) times from concentric Gaussian ellipsoids centred at the current mean (pt) and collect feasible actions based on which cells the samples lie in. [...] We discretized the offer prices and the private valuations to be one of 11 values and set the opportunity cost of losing a user due to low pricing to be 0.5. |