Home

I am Zahra a Ph.D student at Arizona State University, majoring in Computer Science. I work at Yochan research group directed by Prof. Subbarao Kambhampati. My general research interest is to have robots and AIs which are aware of the human in their loop, so they are trustable and explainable. My research interests are Trust, Human-robot interaction, automated planning, and game theory. You can find the latest version of my CV here.

Publications

  • `Trust-Aware Planning: Modeling Trust Evolution in Longitudinal Human-Robot Interaction
  • Z. Zahedi, M. Verma, S. Sreedharan, S. Kambhampati
    XAIP and PlanRob Workshop, ICAPS 2021
  • `Game-theoretic Model of Trust to Infer Human’s Observation Strategy of Robot Behavior
  • S. Sengupta*, Z. Zahedi*, S. Kambhampati
    R4P Workshop, R:SS 2021
  • `Why not give this work to them?' Negotiation-Aware Task-Allocation and Contrastive Explanation Generation
  • Z. Zahedi*, S. Sengupta*, S. Kambhampati
    Cooperative AI Workshop, NeurIPS 2020
  • To Monitor or Not: Observing Robot's Behavior based on a Game-Theoretic Model of Trust
  • S. Sengupta*, Z. Zahedi*, S. Kambhampati
    21st International Workshop on Trust in Agent Societies (co-located with AAMAS), (2019)
  • Towards Understanding User Preferences for Explanation Types in Model Reconciliation
  • Z. Zahedi*, A. Olmo*, T. Chakraborti, S. Sreedharan, S. Kambhampati
    HRI Late Breaking Report, (2019)
  • Fast convergence to Nash equilibria without steady-State oscillation
  • Z. Zahedi, M. M. Arefi, A. Khayatian
    Systems and Control Letters, (2019)
  • Fast seeking of Nash equilibria without steady-state oscillation in games with non-quadratic payoffs
  • Z. Zahedi, M. M. Arefi, A. Khayatian, and H. Modares
    in proc. 2018 American Control Conference
  • Convergence without oscillation to Nash equilibria in non-Cooperative games with quadratic payoffs
  • Z. Zahedi, M. M. Arefi, and A. Khayatian
    25th Iranian Conference on Electrical Engineering (ICEE), published in IEEE Xplorer, May 2017
  • Real-time, Simultaneous Multi-Channel Data Acquisition Systems with no time skews between input channels
  • F. Zahedi, Z. Zahedi
    International Journal of Signal Processing Systems (IJSPS), vol. 4, no. 1, pp. 17-21, 2016
  • A review of Neuro-fuzzy Systems based on Intelligent Control
  • F. Zahedi, Z. Zahedi
    Journal of Electrical and Electronic Engineering, vol. 3, no. 2-1, pp. 58-61, 2015
  • Real-time, Simultaneous Multi-Channel Data Acquisition Systems with no time skews between input channels
  • F. Zahedi, Z. Zahedi
    6th International Conference on Signal Processing Systems (ICSPS), 2014
  • Review of Neuro-fuzzy based on Intelligent Control
  • F. Zahedi, Z. Zahedi
    8th Symposium on Advances in Science & Technology, 2014 (Persian)

    Awards and Honors

    Fulton Fellowship Award

    2019-2020

    Graduate Fellowship Award

    2019 and 2020

    2nd rank among the Control students in Shiraz University

    2016-2017

    Granted merit-based admission to M.Sc. in Shiraz University

    2015

    Honored as Active Student, Shiraz University

    2015

    Current Research Projects

    Trust-Aware Planning in longitudinal human robot interaction - In this work, we have propose a computational model for capturing and modulating trust inlongitudinal human-robot interaction where the robot integrates human’s trust and their expectations into planning to build and maintain trust over the interaction horizon. By establishing the required level of trust, the robot can focus on maximizing the team goal by eschewing explicit explanatory or explicable behavior. The human also with a high level of trust in the robot, might choose not to monitor the robot, or not to intervene by stopping the robot and save their resources.
    Model of trust in monitoring a robot - In this work, we have investigated a human robot interaction scenario when a human supervisor has zero trust on a robot worker by modeling their interaction as Bayesian game. We introduce a notion of trust boundary that optimizes the supervisor’s monitoring cost while ensuring that the robot workers stick to the safe plans.
    Explaining AI-Moderated Task-Allocation Outcomes using Negotiation Trees - We considered a task-allocation problem where an AI Task Allocator (AITA) comes up with a fair allocation for a group of humans. And the AITA is able to provide explanation in the form of negotiation tree to convince humans who thinks a counterfactual allocation is more fair.