Active Search for Sparse Signals with Region Sensing
Authors: Yifei Ma, Roman Garnett, Jeff Schneider
AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the empirical performance of our algorithm on a search problem using satellite image data and in high dimensions. |
| Researcher Affiliation | Academia | Yifei Ma Carnegie Mellon University Pittsburgh PA 15213, US yifeim@cs.cmu.edu Roman Garnett Washington University in St. Louis St. Louis, MO, USA garnett@wustl.edu Jeff Schneider Carnegie Mellon University Pittsburgh PA 15213, US schneide@cs.cmu.edu |
| Pseudocode | Yes | Algorithm 1 Region Sensing Index (RSI) [...] Algorithm 2 Region Sensing Index-Any-k (RSI-A) |
| Open Source Code | No | The paper does not provide any explicit statements or links indicating that source code for the described methodology is publicly available. |
| Open Datasets | Yes | Figure 3 compares the performances on 221 image patches of 512 512 pixels, cropped from National Agriculture Imagery Program (NAIP). [...] https://lta.cr.usgs.gov/node/300 |
| Dataset Splits | No | The paper does not specify training, validation, and test dataset splits. It describes simulations with repetitions and evaluates performance (e.g., recall rate) but does not detail data partitioning for model training or validation. |
| Hardware Specification | No | The paper does not provide any specific hardware details (e.g., CPU, GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, or solvers). |
| Experiment Setup | Yes | Each method was run with 200 repetitions to find its average performance. [...] We picked n = 1024 and various k (sparsity) and d (the dimension of the physical space) annotated below the plots. [...] showing the minimum number of measurements T to guarantee constant Bayes risk ϵT < 0.5. [...] CS then solves a convex optimization problem to infer the nonzero signals, by minimizing t yt x t β 2 2+λ β 1 s.t. β 0, where λ is chosen to produce exactly k nonzero coefficients using the Lasso regularization path. |