Learning Losses for Strategic Classification
Authors: Tosca Lechner, Ruth Urner7337-7344
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In our work we take a learning theoretic perspective, focusing on the sample complexity needed to learn a good decision rule which is robust to strategic manipulation. We perform this analysis by introducing a novel loss function, the strategic manipulation loss, which takes into account both the accuracy of the final decision rule and its vulnerability to manipulation. We analyse the sample complexity for a known graph of possible manipulations in terms of the complexity of the function class and the manipulation graph. Additionally, we initialize the study of learning under unknown manipulation capabilities of the involved agents. Using techniques from transfer learning theory, we define a similarity measure for manipulation graphs and show that learning outcomes are robust with respect to small changes in the manipulation graph. Lastly, we analyse the (sample complexity of) learning of the manipulation capability of agents with respect to this similarity measure, providing novel guarantees for strategic classification with respect to an unknown manipulation graph. |
| Researcher Affiliation | Academia | Tosca Lechner1, Ruth Urner2 1 University of Waterloo, Cheriton School of Computer Science, Waterloo, Canada 2 York University, Lassonde School of Engineering, EECS Department, Toronto, Canada tlechner@uwaterloo.ca, ruth@eecs.yorku.ca |
| Pseudocode | No | The paper is theoretical and focuses on definitions, theorems, and proofs. It does not include any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | No | The paper is theoretical and discusses 'data-generating processes' and 'samples from distribution P' as abstract concepts. It does not use or provide access information for any specific public or open dataset. |
| Dataset Splits | No | The paper is theoretical and does not describe empirical experiments involving dataset splits. Therefore, no training/test/validation split information is provided. |
| Hardware Specification | No | The paper is theoretical and does not mention any hardware used for computations or experiments. |
| Software Dependencies | No | The paper is theoretical and does not mention any specific software or library dependencies with version numbers. |
| Experiment Setup | No | The paper is theoretical and does not describe any empirical experiments, thus no experimental setup details like hyperparameters or training configurations are provided. |