Differentially Private Learning of Geometric Concepts
Authors: Haim Kaplan, Yishay Mansour, Yossi Matias, Uri Stemmer
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | We present differentially private efficient algorithms for learning union of polygons in the plane (which are not necessarily convex). Our algorithms achieve (α, β)-PAC learning and (ε, δ)-differential privacy using a sample of size O 1 / αεk log d , where the domain is [d] [d] and k is the number of edges in the union of polygons. The paper primarily focuses on theoretical aspects, presenting algorithms, theorems, claims, and proofs regarding differential privacy and learning geometric concepts, without detailing any empirical studies, datasets, or experimental results. |
| Researcher Affiliation | Collaboration | 1Tel Aviv University 2Google research, Israel 3Ben-Gurion University 4Supported by a gift from Google Ltd. |
| Pseudocode | Yes | Algorithm Set Cover Learner, Algorithm Select Halfplane |
| Open Source Code | No | The paper states 'Remark 1.5. Any informalities made hereafter are removed in the a full version of the paper, available at https: //arxiv.org/abs/1902.05017.' This link is to the full version of the paper on arXiv, not to source code. No other mention of code availability is found. |
| Open Datasets | No | The paper is theoretical and focuses on algorithm design and proofs, rather than empirical evaluation on datasets. It does not mention the use of any specific publicly available or open datasets for training or evaluation. |
| Dataset Splits | No | The paper is theoretical and does not describe empirical experiments that would involve dataset splits (training, validation, test). Therefore, no dataset split information is provided. |
| Hardware Specification | No | The paper is theoretical and does not describe empirical experiments. As such, no hardware specifications used for running experiments are mentioned. |
| Software Dependencies | No | The paper is theoretical and does not describe any implemented experiments or software that would require specific ancillary software dependencies with version numbers. |
| Experiment Setup | No | The paper is theoretical and focuses on algorithm design and proofs. It does not describe any empirical experimental setup, hyperparameters, or system-level training settings. |