Zobrazeno 1 - 10
of 23
pro vyhledávání: '"Mozannar, Hussein"'
Autor:
Mozannar, Hussein
AI systems are augmenting humans' capabilities in settings such as healthcare and programming, forming human-AI teams. To enable more accurate and timely decisions, we need to optimize the performance of the human-AI team directly. In this thesis, we
Externí odkaz:
https://hdl.handle.net/1721.1/155508
Autor:
Mozannar, Hussein, Chen, Valerie, Alsobay, Mohammed, Das, Subhro, Zhao, Sebastian, Wei, Dennis, Nagireddy, Manish, Sattigeri, Prasanna, Talwalkar, Ameet, Sontag, David
Evaluation of large language models (LLMs) for code has primarily relied on static benchmarks, including HumanEval (Chen et al., 2021), which measure the ability of LLMs to generate complete code that passes unit tests. As LLMs are increasingly used
Externí odkaz:
http://arxiv.org/abs/2404.02806
Autor:
Jeong, Hyewon, Jabbour, Sarah, Yang, Yuzhe, Thapta, Rahul, Mozannar, Hussein, Han, William Jongwon, Mehandru, Nikita, Wornow, Michael, Lialin, Vladislav, Liu, Xin, Lozano, Alejandro, Zhu, Jiacheng, Kocielnik, Rafal Dariusz, Harrigian, Keith, Zhang, Haoran, Lee, Edward, Vukadinovic, Milos, Balagopalan, Aparna, Jeanselme, Vincent, Matton, Katherine, Demirel, Ilker, Fries, Jason, Rashidi, Parisa, Beaulieu-Jones, Brett, Xu, Xuhai Orson, McDermott, Matthew, Naumann, Tristan, Agrawal, Monica, Zitnik, Marinka, Ustun, Berk, Choi, Edward, Yeom, Kristen, Gursoy, Gamze, Ghassemi, Marzyeh, Pierson, Emma, Chen, George, Kanjilal, Sanjat, Oberst, Michael, Zhang, Linying, Singh, Harvineet, Hartvigsen, Tom, Zhou, Helen, Okolo, Chinasa T.
The third ML4H symposium was held in person on December 10, 2023, in New Orleans, Louisiana, USA. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for
Externí odkaz:
http://arxiv.org/abs/2403.01628
Autor:
Mozannar, Hussein
Learning algorithms are often used in conjunction with expert decision makers in practical scenarios, however, this fact is largely ignored when designing these algorithms. In this thesis, we explore how to learn predictors that can either predict or
Externí odkaz:
https://hdl.handle.net/1721.1/151827
Enabling machine learning classifiers to defer their decision to a downstream expert when the expert is more accurate will ensure improved safety and performance. This objective can be achieved with the learning-to-defer framework which aims to joint
Externí odkaz:
http://arxiv.org/abs/2311.01106
Autor:
Mozannar, Hussein, Lee, Jimin J, Wei, Dennis, Sattigeri, Prasanna, Das, Subhro, Sontag, David
People are relying on AI agents to assist them with various tasks. The human must know when to rely on the agent, collaborate with the agent, or ignore its suggestions. In this work, we propose to learn rules, grounded in data regions and described i
Externí odkaz:
http://arxiv.org/abs/2311.01007
AI powered code-recommendation systems, such as Copilot and CodeWhisperer, provide code suggestions inside a programmer's environment (e.g., an IDE) with the aim of improving productivity. We pursue mechanisms for leveraging signals about programmers
Externí odkaz:
http://arxiv.org/abs/2306.04930
Autor:
Mozannar, Hussein, Utsumi, Yuria, Chen, Irene Y., Gervasi, Stephanie S., Ewing, Michele, Smith-McLallen, Aaron, Sontag, David
A high-risk pregnancy is a pregnancy complicated by factors that can adversely affect the outcomes of the mother or the infant. Health insurers use algorithms to identify members who would benefit from additional clinical support. This work presents
Externí odkaz:
http://arxiv.org/abs/2305.17261
Autor:
Mozannar, Hussein, Lang, Hunter, Wei, Dennis, Sattigeri, Prasanna, Das, Subhro, Sontag, David
Automated AI classifiers should be able to defer the prediction to a human decision maker to ensure more accurate predictions. In this work, we jointly train a classifier with a rejector, which decides on each data point whether the classifier or the
Externí odkaz:
http://arxiv.org/abs/2301.06197
Code-recommendation systems, such as Copilot and CodeWhisperer, have the potential to improve programmer productivity by suggesting and auto-completing code. However, to fully realize their potential, we must understand how programmers interact with
Externí odkaz:
http://arxiv.org/abs/2210.14306