Dynamic Sensing: Better Classification under Acquisition Constraints
Authors: Oran Richman, Shie Mannor
ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the potential benefit of this approach on both simulated and real-life problems. and 5. Simulation Results We tested our method on three data-sets. The first is a synthetic toy data-base. The second is the IRIS database from the UCI repository, in this database noise was added artificially. The third is a speaker verification corpus. |
| Researcher Affiliation | Academia | Oran Richman RORAN@TX.TECHNION.AC.IL Department of Electrical Engineering, Technion Israel Institute of Technology, Haifa 32000, Israel Shie Mannor SHIE@EE.TECHNION.AC.IL Department of Electrical Engineering, Technion Israel Institute of Technology, Haifa 32000, Israel |
| Pseudocode | No | The paper contains mathematical derivations and proofs but no pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statements or links indicating that source code for the described methodology is publicly available. |
| Open Datasets | Yes | The second is the IRIS database from the UCI repository (Bache and Lichman, 2013). and We used the MIT Mobile Device Speaker Verification Corpus (Ram Woo and Hazen, 2006). |
| Dataset Splits | No | The paper mentions “training data” and “test phase” but does not explicitly provide details about training, validation, or test dataset splits (e.g., percentages, counts, or explicit validation sets) needed for reproduction. For the IRIS dataset, it describes averaging over noise vectors, not dataset splits. |
| Hardware Specification | No | The paper does not specify any particular hardware (CPU, GPU, etc.) used for running the experiments. |
| Software Dependencies | No | The paper mentions “HTK MFCC matlab package” in a footnote, but does not specify a version number. It also mentions “SVM classification” but not the specific library or its version (e.g., scikit-learn, LIBSVM, etc.). |
| Experiment Setup | Yes | The coarse samples x were generated from the raw samples by adding zero mean Gaussian noise with standard deviation 0.33. Measurement noise was taken to be Gaussian with zero mean and variance [0.33^2+r(x)]^-1... (Toy database setup). coarse acquisition standard deviation of 0.33. (IRIS dataset). The noise was modelled as Gaussian with mean 0 and standard deviation σs^-1 where s is the sampling factor. (Speaker identification). Mel-frequency scale cepstral coefficient (MFCC) are extracted from each recording. This is done using a 25msec window with 10msec interval. (Speaker identification MFCC extraction). |