Top AI Safety papers

Find all the Top AIAI Safety papers. Links to pdf, code repos and demos are provided.

•

almost 5 years ago

Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection

Michael A. Powell, Allison Koenecke, Ruoxuan Xiong + 22 more

In severe viral pneumonia, including Coronavirus disease 2019 (COVID-19), the viral replication phase is often followed by hyperinflammation, which ca...
see more

over 2 years ago

Modelling cortical network dynamics

Gerald K. Cooray, Richard E. Rosch, Karl J. Friston

We consider the theoretical constraints on interactions between coupled cortical columns. Each column comprises a set of neural populations, where eac...
see more

Tabular

almost 3 years ago

Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments

Nicholas J. Wareham, Dimitris Spathis, Ignacio Perez-Pozuelo + 4 more

Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (...
see more

Language

10 months ago

Fairness-Aware Meta-Learning via Nash Bargaining

Michael. I. Jordan, Yi Zeng, Xuelin Yang + 4 more

To address issues of group-level fairness in machine learning, it is natural to adjust model parameters based on specific fairness objectives over a s...
see more

Language

about 1 year ago

Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference

Michael. I. Jordan, Wei-Lin Chiang, Lianmin Zheng + 8 more

Introduces Chatbot Arena, an innovative open platform designed for evaluating large language models (LLMs) based on human preferences through a crowds...
see more

about 1 year ago

On Three-Layer Data Markets

Michael. I. Jordan, Alireza Fallah, Ali Makhdoumi + 1 more

We study a three-layer data market comprising users (data owners), platforms, and a data buyer. Each user benefits from platform services in exchange ...
see more

about 1 year ago

The Limits of Price Discrimination Under Privacy Constraints

Michael. I. Jordan, Alireza Fallah, Ali Makhdoumi + 1 more

We study a producer's problem of selling a product to a continuum of privacy-conscious consumers, where the producer can implement third-degree price ...
see more

Language

about 1 year ago

Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF

Michael. I. Jordan, Banghua Zhu, Jiantao Jiao

Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique that aligns language models closely with human-centric values. The initial ph...
see more

over 1 year ago

Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions

Michael. I. Jordan, Anastasios Nikolas Angelopoulos, Jordan Lekeufack + 2 more

We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples o...
see more

over 1 year ago

Scaff-PD: Communication Efficient Fair and Robust Federated Learning

Michael. I. Jordan, Yaodong Yu, Sai Praneeth Karimireddy + 1 more

We present Scaff-PD, a fast and communication-efficient algorithm for distributionally robust federated learning. Our approach improves fairness by op...
see more

almost 2 years ago

Incentivizing High-Quality Content in Online Recommender Systems

Michael. I. Jordan, Xinyan Hu, Meena Jagadeesan + 1 more

In content recommender systems such as TikTok and YouTube, the platform's recommendation algorithm shapes content producer incentives. Many platforms ...
see more

#12

Computer Vision

almost 2 years ago

Federated Conformal Predictors for Distributed Uncertainty Quantification

Michael. I. Jordan, Charles Lu, Yaodong Yu + 2 more

Conformal prediction is emerging as a popular paradigm for providing rigorous uncertainty quantification in machine learning since it can be easily ap...
see more

Loading. Please wait...

Top AI Safety papers

Find all the Top AIAI Safety papers. Links to pdf, code repos and demos are provided.

•

almost 5 years ago

Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection

Michael A. Powell, Allison Koenecke, Ruoxuan Xiong + 22 more

In severe viral pneumonia, including Coronavirus disease 2019 (COVID-19), the viral replication phase is often followed by hyperinflammation, which ca...
see more

over 2 years ago

Modelling cortical network dynamics

Gerald K. Cooray, Richard E. Rosch, Karl J. Friston

We consider the theoretical constraints on interactions between coupled cortical columns. Each column comprises a set of neural populations, where eac...
see more

Tabular

almost 3 years ago

Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments

Nicholas J. Wareham, Dimitris Spathis, Ignacio Perez-Pozuelo + 4 more

Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (...
see more

Language

10 months ago

Fairness-Aware Meta-Learning via Nash Bargaining

Michael. I. Jordan, Yi Zeng, Xuelin Yang + 4 more

To address issues of group-level fairness in machine learning, it is natural to adjust model parameters based on specific fairness objectives over a s...
see more

Language

about 1 year ago

Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference

Michael. I. Jordan, Wei-Lin Chiang, Lianmin Zheng + 8 more

Introduces Chatbot Arena, an innovative open platform designed for evaluating large language models (LLMs) based on human preferences through a crowds...
see more

about 1 year ago

On Three-Layer Data Markets

Michael. I. Jordan, Alireza Fallah, Ali Makhdoumi + 1 more

We study a three-layer data market comprising users (data owners), platforms, and a data buyer. Each user benefits from platform services in exchange ...
see more

about 1 year ago

The Limits of Price Discrimination Under Privacy Constraints

Michael. I. Jordan, Alireza Fallah, Ali Makhdoumi + 1 more

We study a producer's problem of selling a product to a continuum of privacy-conscious consumers, where the producer can implement third-degree price ...
see more

Language

about 1 year ago

Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF

Michael. I. Jordan, Banghua Zhu, Jiantao Jiao

Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique that aligns language models closely with human-centric values. The initial ph...
see more

over 1 year ago

Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions

Michael. I. Jordan, Anastasios Nikolas Angelopoulos, Jordan Lekeufack + 2 more

We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples o...
see more

over 1 year ago

Scaff-PD: Communication Efficient Fair and Robust Federated Learning

Michael. I. Jordan, Yaodong Yu, Sai Praneeth Karimireddy + 1 more

We present Scaff-PD, a fast and communication-efficient algorithm for distributionally robust federated learning. Our approach improves fairness by op...
see more

almost 2 years ago

Incentivizing High-Quality Content in Online Recommender Systems

Michael. I. Jordan, Xinyan Hu, Meena Jagadeesan + 1 more

In content recommender systems such as TikTok and YouTube, the platform's recommendation algorithm shapes content producer incentives. Many platforms ...
see more

#12

Computer Vision

almost 2 years ago

Federated Conformal Predictors for Distributed Uncertainty Quantification

Michael. I. Jordan, Charles Lu, Yaodong Yu + 2 more

Conformal prediction is emerging as a popular paradigm for providing rigorous uncertainty quantification in machine learning since it can be easily ap...
see more