Research

Research Interests:
Human-in-the-loop control systems
Statistical learning
Vehicle dynamics and control
Engineering design

Research summary:
I study how humans and automated systems can work together to complete physical tasks. Often these tasks are expressed as optimization problems. The humans interact with the automated system by defining the objective function to be optimized, by providing information about the structure of the problem, and sometimes by approving and implementing the actions recommended by the automated system. By combining the ability of humans to learn structure from experience and the ability of automated algorithms to carry out computation and process data, I develop systems that achieve higher performance than either humans or automation could achieve on their own.

 

Models of human search behavior

Search is the process of making rational decisions under uncertainty. In [1] we developed a model of human behavior in a spatial search task. Notably, we found that some people achieved very high performance on such tasks, including significant numbers of people who achieved performance better than an otherwise “optimal” algorithm. Our model allowed us to show that this high performance is consistent with the humans having good priors, i.e., the high-performing people have good intuitions about the task. In [2], we considered the problem of learning these intuitions from data and showed that this could be achieved using our model. In [3], we have considered other, more sophisticated models of human choice behavior based on the concept of satisficing, where the decision maker cares only about achieving performance above a given threshold, rather than achieving maximum possible performance.

[1] [pdf] P. B. Reverdy, V. Srivastava, and N. E. Leonard, “Modeling human decision making in generalized Gaussian multiarmed bandits,” Proceedings of the IEEE, vol. 102, iss. 4, p. 544–571, 2014.
[Bibtex]
@article{PBR-VS-NEL:14,
Author = {Reverdy, Paul B and Srivastava, Vaibhav and Leonard, Naomi Ehrich},
Date-Added = {2018-02-05 06:22:16 +0000},
Date-Modified = {2018-02-08 21:11:54 +0000},
Journal = {Proceedings of the {IEEE}},
Number = {4},
Pages = {544--571},
Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PBR-VS-NEL-14.pdf},
Publisher = {IEEE},
Title = {Modeling human decision making in generalized {G}aussian multiarmed bandits},
Volume = {102},
Year = {2014}}
[2] [pdf] [doi] P. Reverdy and N. E. Leonard, “Parameter estimation in softmax decision-making models with linear objective functions,” IEEE Transactions on Automation Science and Engineering, vol. 13, iss. 1, p. 54–67, 2016.
[Bibtex]
@article{PR-NEL:16,
Author = {Reverdy, Paul and Leonard, Naomi Ehrich},
Date-Added = {2018-02-05 06:22:16 +0000},
Date-Modified = {2018-02-08 21:37:50 +0000},
Doi = {10.1109/TASE.2015.2499244},
Journal = {{IEEE} {T}ransactions on {A}utomation {S}cience and {E}ngineering},
Number = {1},
Pages = {54--67},
Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PR-NEL-16.pdf},
Publisher = {IEEE},
Title = {Parameter estimation in softmax decision-making models with linear objective functions},
Volume = {13},
Year = {2016},
Bdsk-Url-1 = {https://doi.org/10.1109/TASE.2015.2499244}}
[3] [doi] P. Reverdy, V. Srivastava, and N. E. Leonard, “Satisficing in multi-armed bandit problems,” IEEE Transactions on Automatic Control, vol. 62, iss. 8, p. 3788–3803, 2017.
[Bibtex]
@article{PR-VS-NEL:17,
Author = {Reverdy, Paul and Srivastava, Vaibhav and Leonard, Naomi Ehrich},
Date-Added = {2018-02-05 06:22:16 +0000},
Date-Modified = {2018-02-08 21:34:38 +0000},
Doi = {10.1109/TAC.2016.2644380},
Journal = {{IEEE} {T}ransactions on {A}utomatic {C}ontrol},
Number = {8},
Pages = {3788--3803},
Publisher = {IEEE},
Title = {Satisficing in multi-armed bandit problems},
Volume = {62},
Year = {2017},
Bdsk-Url-1 = {https://doi.org/10.1109/TAC.2016.2644380}}
  • [PDF] P. B. Reverdy, V. Srivastava, and N. E. Leonard, “Modeling human decision making in generalized Gaussian multiarmed bandits,” Proceedings of the IEEE, vol. 102, iss. 4, p. 544–571, 2014.
    [Bibtex]
    @article{PBR-VS-NEL:14,
    Author = {Reverdy, Paul B and Srivastava, Vaibhav and Leonard, Naomi Ehrich},
    Date-Added = {2018-02-05 06:22:16 +0000},
    Date-Modified = {2018-02-08 21:11:54 +0000},
    Journal = {Proceedings of the {IEEE}},
    Number = {4},
    Pages = {544--571},
    Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PBR-VS-NEL-14.pdf},
    Publisher = {IEEE},
    Title = {Modeling human decision making in generalized {G}aussian multiarmed bandits},
    Volume = {102},
    Year = {2014}}

Reactive decision making

  • [PDF] P. B. Reverdy and D. E. Koditschek, “A dynamical system for prioritizing and coordinating motivations,” Arxiv preprint arxiv:1703.01662, 2017.
    [Bibtex]
    @article{PBR-DEK:17,
    Author = {Reverdy, Paul B and Koditschek, Daniel E},
    Date-Added = {2018-10-10 19:24:34 -0700},
    Date-Modified = {2018-10-10 19:24:34 -0700},
    Journal = {arXiv preprint arXiv:1703.01662},
    Pdf = {https://arxiv.org/abs/1703.01662},
    Title = {A dynamical system for prioritizing and coordinating motivations},
    Url = {https://arxiv.org/abs/1703.01662},
    Year = {2017},
    Bdsk-Url-1 = {https://arxiv.org/abs/1703.01662}}
  • [PDF] P. Reverdy, D. B. Ilhan, and D. E. Koditschek, “A drift-diffusion model for robotic obstacle avoidance,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2015, p. 6113–6120.
    [Bibtex]
    @inproceedings{PR-BDI-DEK:15,
    Author = {Reverdy, Paul and Ilhan, B Deniz and Koditschek, Daniel E},
    Booktitle = {{IEEE/RSJ} {I}nternational {C}onference on {I}ntelligent {R}obots and {S}ystems ({IROS})},
    Date-Added = {2018-02-05 06:22:16 +0000},
    Date-Modified = {2018-02-08 21:16:42 +0000},
    Organization = {IEEE},
    Pages = {6113--6120},
    Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PR-BDI-DEK-15.pdf},
    Title = {A drift-diffusion model for robotic obstacle avoidance},
    Year = {2015}}
  • [PDF] P. Reverdy and D. E. Koditschek, “Mobile robots as remote sensors for spatial point process models,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2016, p. 2847–2852.
    [Bibtex]
    @inproceedings{PR-DEK:16,
    Author = {Reverdy, Paul and Koditschek, Daniel E},
    Booktitle = {{IEEE/RSJ} {I}nternational {C}onference on {I}ntelligent {R}obots and {S}ystems ({IROS})},
    Date-Added = {2018-02-05 06:22:16 +0000},
    Date-Modified = {2018-02-08 21:20:07 +0000},
    Organization = {IEEE},
    Pages = {2847--2852},
    Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PR-DEK-16.pdf},
    Title = {Mobile robots as remote sensors for spatial point process models},
    Year = {2016}}

Engineering design

  • [PDF] P. Reverdy, A. Reddy, L. Martinelli, and N. E. Leonard, “Integrating a human designer’s preferences in Multidisciplinary Design Optimization,” in 15th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, 2014.
    [Bibtex]
    @inproceedings{PR-etal:14,
    Author = {Reverdy, Paul and Reddy, Akhil and Martinelli, Luigi and Leonard, Naomi E},
    Booktitle = {15th {AIAA/ISSMO} {M}ultidisciplinary {A}nalysis and {O}ptimization {C}onference},
    Date-Added = {2018-02-05 06:22:16 +0000},
    Date-Modified = {2018-02-08 21:11:10 +0000},
    Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PR-etal-14.pdf},
    Title = {Integrating a human designer's preferences in {M}ultidisciplinary {D}esign {O}ptimization},
    Year = {2014}}
  • [PDF] P. Reverdy, “Gaussian multi-armed bandit problems with multiple objectives,” in American Control Conference (ACC), 2016, 2016, p. 5263–5269.
    [Bibtex]
    @inproceedings{PR:16,
    Author = {Reverdy, Paul},
    Booktitle = {American {C}ontrol {C}onference ({ACC}), 2016},
    Date-Added = {2018-02-05 06:22:16 +0000},
    Date-Modified = {2018-02-08 21:18:06 +0000},
    Organization = {IEEE},
    Pages = {5263--5269},
    Pdf = {http://www.paulreverdy.com/wp-content/papercite-data/pdf/PR-16.pdf},
    Title = {Gaussian multi-armed bandit problems with multiple objectives},
    Year = {2016}}