CAREER: Towards a theory of machine learning with strategic interactions
职业:走向具有战略互动的机器学习理论
基本信息
- 批准号:2145898
- 负责人:
- 金额:$ 60万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Continuing Grant
- 财政年份:2022
- 资助国家:美国
- 起止时间:2022-03-01 至 2027-02-28
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Machine learning (ML) algorithms use observed sampled data to uncover general patterns that can then be used for making predictions. Learning systems that interact with human data and stakeholders (such as those used in personalized medicine, content curation, financial markets, hiring, and lending) take place in a complex social and economic context. In this wide range of applications, there are feedback loops between learning algorithms and people that impact the quality of the learning process and the wellbeing of people. These feedback loops are currently not captured by the classical theory of ML, and handling them in an ad hoc, non-mathematical, way could have major social repercussions. This project will develop a rigorous mathematical framework for addressing interactions between learning systems and people and will draw from a wide range of academic traditions and fields, including Theory of Computing, Artificial Intelligence, Economics and Computation. This project also addresses educational and community building plans for enabling the next generations of students to contribute to a theory of machine learning for emerging and modern needs through cross-disciplinary research.This project will build a theoretical foundation for ensuring both the performance of learning algorithms in the presence of everyday social and economic forces and the integrity of social and economic forces that are born out of the use of machine-learning systems. To achieve this, the investigator will consider adversarial, strategic, and collaborative interactions. For adversarial and long-term strategic interactions, the project will explore online decision processes and contribute online learning algorithms that perform well in presence of more realistic adaptive and non-myopic strategic agents. The project also explores the long-term social impact of strategic play and communication on learning and quality of available information, with an eye towards understanding and addressing biased and polarized beliefs. Additionally, to reap the full benefit of collaborative interactions, the project will align the performance of learning algorithms with the needs and preferences of participating agents. This will lead to the design of collaborative learning protocols that are differentially private, statistically efficient, and equitable. This project also includes outreach, mentoring, and educational plans that will complement its technical goals, including a workshop series on "Learning in presence of Strategic Behavior" that brings together members of different communities and helps set an agenda for the field and "Learning Theory Alliance" that is a large-scale mentoring initiative for supporting the machine learning theory community.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
该奖项全部或部分由2021年美国救援计划法案(公法117-2)资助。机器学习(ML)算法使用观察到的采样数据来揭示一般模式,然后可以用于进行预测。与人类数据和利益相关者交互的学习系统(例如用于个性化医疗,内容策展,金融市场,招聘和贷款的系统)发生在复杂的社会和经济背景下。在这种广泛的应用中,学习算法和人之间存在反馈回路,影响学习过程的质量和人们的福祉。这些反馈循环目前还没有被ML的经典理论所捕获,以一种特别的、非数学的方式处理它们可能会产生重大的社会影响。该项目将开发一个严格的数学框架,用于解决学习系统与人之间的交互,并将借鉴广泛的学术传统和领域,包括计算理论,人工智能,经济学和计算。该项目还涉及教育和社区建设计划,使下一代学生能够通过跨学科和跨学科的方法,为新兴和现代需求的机器学习理论做出贡献。该项目将建立一个理论基础,以确保在日常社会和经济力量存在的情况下学习算法的性能,以及从使用中产生的社会和经济力量的完整性。机器学习系统。为了实现这一目标,研究者将考虑对抗性、战略性和协作性的互动。对于对抗性和长期的战略互动,该项目将探索在线决策过程,并提供在线学习算法,这些算法在更现实的适应性和非短视的战略代理中表现良好。该项目还探讨了战略游戏和沟通对学习和现有信息质量的长期社会影响,着眼于理解和解决偏见和两极分化的信念。此外,为了获得协作交互的全部好处,该项目将使学习算法的性能与参与代理的需求和偏好保持一致。这将导致合作学习协议的设计,是不同的私人,统计效率和公平。该项目还包括推广、指导和教育计划,以补充其技术目标,包括一个关于“战略行为存在下的学习”的系列研讨会,该研讨会汇集了不同社区的成员,并有助于为该领域和“学习理论联盟”制定议程,这是一个大型的-该奖项反映了NSF的法定使命,并被认为值得通过使用基金会的学术价值和更广泛的影响审查标准。
项目成果
期刊论文数量(15)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Oracle-Efficient Online Learning for Smoothed Adversaries
Oracle 高效在线学习,轻松应对对手
- DOI:
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Haghtalab, Nika;Han, Yanjun;Shetty, Abhishek;Yang, Kunhe
- 通讯作者:Yang, Kunhe
Leveraging Reviews: Learning to Price with Buyer and Seller Uncertainty
利用评论:学习在买家和卖家的不确定性下定价
- DOI:10.1145/3580507.3597663
- 发表时间:2023
- 期刊:
- 影响因子:0
- 作者:Guo, Wenshuo;Haghtalab, Nika;Kandasamy, Kirthevasan;Vitercik, Ellen
- 通讯作者:Vitercik, Ellen
A Unifying Perspective on Multi-Calibration: Game Dynamics for Multi-Objective Learning
多重校准的统一视角:多目标学习的游戏动力学
- DOI:
- 发表时间:2023
- 期刊:
- 影响因子:0
- 作者:Haghtalab, Nika;Jordan, Michael;Zhao, Eric
- 通讯作者:Zhao, Eric
Smoothed Analysis of Sequential Probability Assignment
顺序概率分配的平滑分析
- DOI:
- 发表时间:2023
- 期刊:
- 影响因子:0
- 作者:Bhatt, Alankrita;Haghtalab, Nika;Shetty, Abhishek
- 通讯作者:Shetty, Abhishek
Learning in Stackelberg Games with Non-myopic Agents
与非近视智能体一起在 Stackelberg 游戏中学习
- DOI:10.1145/3490486.3538308
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Haghtalab, Nika;Lykouris, Thodoris;Nietert, Sloan;Wei, Alexander
- 通讯作者:Wei, Alexander
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Nika Haghtalab其他文献
Monitoring stealthy diffusion
- DOI:
10.1007/s10115-017-1023-7 - 发表时间:
2017-02-13 - 期刊:
- 影响因子:3.100
- 作者:
Nika Haghtalab;Aron Laszka;Ariel D. Procaccia;Yevgeniy Vorobeychik;Xenofon Koutsoukos - 通讯作者:
Xenofon Koutsoukos
Theory of Computing
计算理论
- DOI:
10.4086/toc - 发表时间:
2013 - 期刊:
- 影响因子:0
- 作者:
Alexandr Andoni;Nikhil Bansal;P. Beame;Giuseppe Italiano;Sanjeev Khanna;Ryan O’Donnell;T. Pitassi;T. Rabin;Tim Roughgarden;Clifford Stein;Rocco Servedio;Amir Abboud;Nima Anari;Ibm Srinivasan Arunachalam;T. J. Watson;Research Center;Petra Berenbrink;Aaron Bernstein;Aditya Bhaskara;Sayan Bhattacharya;Eric Blais;H. Bodlaender;Adam Bouland;Anne Broadbent;Mark Bun;Timothy Chan;Arkadev Chattopadhyay;Xue Chen;Gil Cohen;Dana Dachman;Anindya De;Shahar Dobzhinski;Zhiyi Huang;Ken;Robin Kothari;Marvin Künnemann;Tu Kaiserslautern;Rasmus Kyng;E. Zurich;Sophie Laplante;D. Lokshtanov;S. Mahabadi;Nicole Megow;Ankur Moitra;Technion Shay Moran;Google Research;Christopher Musco;Prasad Raghavendra;Alex Russell;Laura Sanità;Alex Slivkins;David Steurer;Epfl Ola Svensson;Chaitanya Swamy;Madhur Tulsiani;Christos Tzamos;Andreas Wiese;Mary Wootters;Huacheng Yu;Aaron Potechin;Aaron Sidford;Aarushi Goel;Aayush Jain;Abhiram Natarajan;Abhishek Shetty;Adam Karczmarz;Adam O’Neill;Aditi Dudeja;Aditi Laddha;Aditya Krishnan;Adrian Vladu Afrouz;J. Ameli;Ainesh Bakshi;Akihito Soeda;Akshay Krishnamurthy;Albert Cheu;A. Grilo;Alex Wein;Alexander Belov;Alexander Block;Alexander Golovnev;Alexander Poremba;Alexander Shen;Alexander Skopalik;Alexandra Henzinger;Alexandros Hollender;Ali Parviz;Alkis Kalavasis;Allen Liu;Aloni Cohen;Amartya Shankha;Biswas Amey;Bhangale Amin;Coja;Yehudayoff Amir;Zandieh Amit;Daniely Amit;Kumar Amnon;Ta;Beimel Anand;Louis Anand Natarajan;Anders Claesson;André Chailloux;André Nusser;Andrea Coladangelo;Andrea Lincoln;Andreas Björklund;Andreas Maggiori;A. Krokhin;A. Romashchenko;Andrej Risteski;Anirban Chowdhury;Anirudh Krishna;A. Mukherjee;Ankit Garg;Anna Karlin;Anthony Leverrier;Antonio Blanca;A. Antoniadis;Anupam Gupta;Anupam Prakash;A. Singh;Aravindan Vijayaraghavan;Argyrios Deligkas;Ariel Kulik;Ariel Schvartzman;Ariel Shaulker;A. Cornelissen;Arka Rai;Choudhuri Arkady;Yerukhimovich Arnab;Bhattacharyya Arthur Mehta;Artur Czumaj;A. Backurs;A. Jambulapati;Ashley Montanaro;A. Sah;A. Mantri;Aviad Rubinstein;Avishay Tal;Badih Ghazi;Bartek Blaszczyszyn;Benjamin Moseley;Benny Pinkas;Bento Natura;Bernhard Haeupler;Bill Fefferman;B. Mance;Binghui Peng;Bingkai Lin;B. Sinaimeri;Bo Waggoner;Bodo Manthey;Bohdan Kivva;Brendan Lucier Bundit;Laekhanukit Burak;Sahinoglu Cameron;Seth Chaodong Zheng;Charles Carlson;Chen;Chenghao Guo;Chenglin Fan;Chenwei Wu;Chethan Kamath;Chi Jin;J. Thaler;Jyun;Kaave Hosseini;Kaito Fujii;Kamesh Munagala;Kangning Wang;Kanstantsin Pashkovich;Karl Bringmann Karol;Wegrzycki Karteek;Sreenivasaiah Karthik;Chandrasekaran Karthik;Sankararaman Karthik;C. S. K. Green;Larsen Kasturi;Varadarajan Keita;Xagawa Kent Quanrud;Kevin Schewior;Kevin Tian;Kilian Risse;Kirankumar Shiragur;K. Pruhs;K. Efremenko;Konstantin Makarychev;Konstantin Zabarnyi;Krišj¯anis Pr¯usis;Kuan Cheng;Kuikui Liu;Kunal Marwaha;Lars Rohwedder László;Kozma László;A. Végh;L'eo Colisson;Leo de Castro;Leonid Barenboim Letong;Li;Li;L. Roditty;Lieven De;Lathauwer Lijie;Chen Lior;Eldar Lior;Rotem Luca Zanetti;Luisa Sinisclachi;Luke Postle;Luowen Qian;Lydia Zakynthinou;Mahbod Majid;Makrand Sinha;Malin Rau Manas;Jyoti Kashyop;Manolis Zampetakis;Maoyuan Song;Marc Roth;Marc Vinyals;Marcin Bieńkowski;Marcin Pilipczuk;Marco Molinaro;Marcus Michelen;Mark de Berg;M. Jerrum;Mark Sellke;Mark Zhandry;Markus Bläser;Markus Lohrey;Marshall Ball;Marthe Bonamy;Martin Fürer;Martin Hoefer;M. Kokainis;Masahiro Hachimori;Matteo Castiglioni;Matthias Englert;Matti Karppa;Max Hahn;Max Hopkins;Maximilian Probst;Gutenberg Mayank Goswami;Mehtaab Sawhney;Meike Hatzel;Meng He;Mengxiao Zhang;Meni Sadigurski;M. Parter;M. Dinitz;Michael Elkin;Michael Kapralov;Michael Kearns;James R. Lee;Sudatta Bhattacharya;Michal Koucký;Hadley Black;Deeparnab Chakrabarty;C. Seshadhri;Mahsa Derakhshan;Naveen Durvasula;Nika Haghtalab;Peter Kiss;Thatchaphol Saranurak;Soheil Behnezhad;M. Roghani;Hung Le;Shay Solomon;Václav Rozhon;Anders Martinsson;Christoph Grunau;G. Z. —. Eth;Zurich;Switzerland;Morris Yau — Massachusetts;Noah Golowich;Dhruv Rohatgi — Massachusetts;Qinghua Liu;Praneeth Netrapalli;Csaba Szepesvári;Debarati Das;Jacob Gilbert;Mohammadtaghi Hajiaghayi;Tomasz Kociumaka;B. Saha;K. Bringmann;Nick Fischer — Weizmann;Ce Jin;Yinzhan Xu — Massachusetts;Virginia Vassilevska Williams;Yinzhan Xu;Josh Alman;Kevin Rao;Hamed Hatami;—. XiangMeng;McGill University;Edith Cohen;Xin Lyu;Tamás Jelani Nelson;Uri Stemmer — Google;Research;Daniel Alabi;Pravesh K. Kothari;Pranay Tankala;Prayaag Venkat;Fred Zhang;Samuel B. Hopkins;Gautam Kamath;Shyam Narayanan — Massachusetts;Marco Gaboardi;R. Impagliazzo;Rex Lei;Satchit Sivakumar;Jessica Sorrell;T. Korhonen;Marco Bressan;Matthias Lanzinger;Huck Bennett;Mahdi Cheraghchi;V. Guruswami;João Ribeiro;Jan Dreier;Nikolas Mählmann;Sebastian Siebertz — TU Wien;The Randomized k ;Conjecture Is;False;Sébastien Bubeck;Christian Coester;Yuval Rabani — Microsoft;Wei;Ethan Mook;Daniel Wichs;Joshua Brakensiek;Sai Sandeep — Stanford;University;Lorenzo Ciardo;Stanislav Živný;Amey Bhangale;Subhash Khot;Dor Minzer;David Ellis;Guy Kindler;Noam Lifshitz;Ronen Eldan;Dan Mikulincer;George Christodoulou;E. Koutsoupias;Annamária Kovács;José Correa;Andrés Cristi;Xi Chen;Matheus Venturyne;Xavier Ferreira;David C. Parkes;Yang Cai;Jinzhao Wu;Zhengyang Liu;Zeyu Ren;Zihe Wang;Ravishankar Krishnaswamy;Shi Li;Varun Suriyanarayana - 通讯作者:
Varun Suriyanarayana
A Unifying Perspective on Multi-Calibration: Unleashing Game Dynamics for Multi-Objective Learning
多重校准的统一视角:释放多目标学习的游戏动力
- DOI:
10.48550/arxiv.2302.10863 - 发表时间:
2023 - 期刊:
- 影响因子:0
- 作者:
Nika Haghtalab;Michael I. Jordan;Eric Zhao - 通讯作者:
Eric Zhao
Polarization Through the Lens of Learning Theory
学习理论视角下的两极分化
- DOI:
- 发表时间:
2019 - 期刊:
- 影响因子:0
- 作者:
Nika Haghtalab;M. Jackson;Ariel D. Procaccia - 通讯作者:
Ariel D. Procaccia
Nika Haghtalab的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
相似海外基金
CAREER: Towards a Theory of Deep Learning
职业:走向深度学习理论
- 批准号:
2144994 - 财政年份:2022
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Towards a Theory of Robust Learning & Control for Safety-Critical Autonomous Systems
职业生涯:迈向稳健学习理论
- 批准号:
2045834 - 财政年份:2021
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Towards a Predictive Theory of Algorithmic Mechanism Design
职业:算法机制设计的预测理论
- 批准号:
1942497 - 财政年份:2020
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Towards a Robust Theory of Mechanism Design
职业生涯:建立稳健的机构设计理论
- 批准号:
1942583 - 财政年份:2020
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: First-Principles Predictive Theory and Microscopic Understanding of Nonlinear Light-Matter Interactions towards Designer Nonlinear Optical Materials
职业:设计非线性光学材料的非线性光与物质相互作用的第一原理预测理论和微观理解
- 批准号:
1753054 - 财政年份:2018
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Strategic Advancement of Chemical Theory and Computations towards Complex Synthetic Transformations
职业:化学理论和计算在复杂合成转化方面的战略进步
- 批准号:
1352663 - 财政年份:2014
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Towards an accurate and illuminating theory of weak interactions between open-shell systems
职业:建立一个准确且富有启发性的开壳系统弱相互作用理论
- 批准号:
1351978 - 财政年份:2014
- 资助金额:
$ 60万 - 项目类别:
Standard Grant
CAREER: Towards HCI Theory for Technical and Gender Identity
职业生涯:迈向技术和性别认同的人机交互理论
- 批准号:
1253465 - 财政年份:2013
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant
CAREER: Towards An Optimization-Based and Experimentally Verified Predictive Theory of Human Locomotion
职业:建立基于优化且经过实验验证的人类运动预测理论
- 批准号:
1254842 - 财政年份:2013
- 资助金额:
$ 60万 - 项目类别:
Standard Grant
CAREER: Towards a Formal Theory of Wireless Networking
职业:走向无线网络的正式理论
- 批准号:
0952867 - 财政年份:2010
- 资助金额:
$ 60万 - 项目类别:
Continuing Grant