Diameter-Based Active Learning

Authors: Christopher Tosh, Sanjoy Dasgupta

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We provide, for the first time, an efficient algorithm that is able to realize this upper bound, and we empirically demonstrate its good performance.
Researcher Affiliation Academia 1Department of Computer Science and Engineering, UC San Diego, La Jolla, CA, USA.
Pseudocode Yes Algorithm 1 DBAL; Algorithm 2 SELECT
Open Source Code No The paper does not include any statement or link providing access to open-source code for the methodology described.
Open Datasets No The paper describes generating data for simulations (e.g., 'both the prior distribution and the data distribution are uniform over the unit sphere'), rather than using pre-existing publicly available datasets with access information.
Dataset Splits No The paper describes simulation setups and data generation but does not specify training, validation, or test dataset splits in percentages or sample counts, nor does it reference predefined splits.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python 3.8, PyTorch 1.9).
Experiment Setup Yes We tested on two hypothesis classes: homogeneous, or through-the-origin, linear separators and k-sparse monotone disjunctions. In our simulations, both the prior distribution and the data distribution are uniform over the unit sphere. In our simulations, each data point is a vector whose coordinates are i.i.d. Bernoulli random variables with parameter p. Figure 1 shows the results of our simulations on homogeneous linear separators. Left: d = 10. Middle: d = 25. Right: d = 50. The results of our simulations on k-sparse monotone disjunctions are in Figure 2. In all cases k = 4. Top left: d = 75, p = 0.25. Top right: d = 75, p = 0.5. Bottom left: d = 100, p = 0.25. Bottom right: d = 100, p = 0.5.