Výsledky vyhledávání - "Snoswell, Aaron J"

Report

Measuring Misogyny in Natural Language Generation: Preliminary Results from a Case Study on two Reddit Communities

Autor: Snoswell, Aaron J., Nelson, Lucinda, Xue, Hao, Salim, Flora D., Suzor, Nicolas, Burgess, Jean

Generic `toxicity' classifiers continue to be used for evaluating the potential for harm in natural language generation, despite mounting evidence of their shortcomings. We consider the challenge of measuring misogyny in natural language generation,

Externí odkaz: http://arxiv.org/abs/2312.03330

Zobrazit plný text záznamu

Report

Sociotechnical Specification for the Broader Impacts of Autonomous Vehicles

Autor: Gilbert, Thomas Krendl, Snoswell, Aaron J., Dennis, Michael, McAllister, Rowan, Wu, Cathy

Autonomous Vehicles (AVs) will have a transformative impact on society. Beyond the local safety and efficiency of individual vehicles, these effects will also change how people interact with the entire transportation system. This will generate a dive

Externí odkaz: http://arxiv.org/abs/2205.07395

Zobrazit plný text záznamu

Report

LiMIIRL: Lightweight Multiple-Intent Inverse Reinforcement Learning

Autor: Snoswell, Aaron J., Singh, Surya P. N., Ye, Nan

Multiple-Intent Inverse Reinforcement Learning (MI-IRL) seeks to find a reward function ensemble to rationalize demonstrations of different but unlabelled intents. Within the popular expectation maximization (EM) framework for learning probabilistic

Externí odkaz: http://arxiv.org/abs/2106.01777

Zobrazit plný text záznamu

Report

Revisiting Maximum Entropy Inverse Reinforcement Learning: New Perspectives and Algorithms

Autor: Snoswell, Aaron J., Singh, Surya P. N., Ye, Nan

We provide new perspectives and inference algorithms for Maximum Entropy (MaxEnt) Inverse Reinforcement Learning (IRL), which provides a principled method to find a most non-committal reward function consistent with given expert demonstrations, among

Externí odkaz: http://arxiv.org/abs/2012.00889

Zobrazit plný text záznamu