Information gathering actions over human internal state

By Dorsa Sadigh, S. Sastry, Sanjit A. Seshia, Anca Dragan. 2016

204 citations recorded by the source corpus.

Subfields: Automotive Engineering, Computer Vision and Pattern Recognition, Social Psychology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Automotive Engineering

In Computer Vision and Pattern Recognition

In Social Psychology

Citations in the corpus

Verifiable Reinforcement Learning via Policy Extraction (2018, background): “For example, self-driving cars must robustly handle a variety of human behaviors [26], controllers for robotics typically need stability guarantees [2, 20, 8], and air traffic control should provably satisfy safety properties including robustness [19].”
Autonomous agents modelling other agents: A comprehensive survey and open problems (2017, methodology): “[234] use a form of model predictive control to optimise a heuristic tradeoff between minimising uncertainty in the modelled agent’s type and maximising a given reward function.”
Planning and Decision-Making for Autonomous Vehicles (2018, background): “An extension (108) gathers information about the internal state of another vehicle’s driver by including the information gain over a belief state in the objective function, effectively reducing the entropy.”
Shared Autonomy with Learned Latent Actions (2020, background): “Several of these works infer human intent by maintaining a belief over the set of possible goals, and using human inputs as evidence in a Bayesian framework to continually update this belief [8, 16, 22, 29, 30].”
Machine Teaching for Inverse Reinforcement Learning: Algorithms and Applications (2018, background): “…Q-value of a state-action pair ( s, a ) is deﬁned as and we denote the optimal Q-value function as As is common in the literature (Ziebart et al. 2008; Sadigh et al. 2016; Pirotta and Restelli 2016; Barreto et al. 2017), we assume that the reward function can be expressed as a linear…”
Who is Mistaken? (2016, background): “ral key applications in computer vision, such as in action understanding, robotics, and healthcare. For example, understanding beliefs of human drivers could improve the safety of autonomous vehicles [25]. Robots that understand human beliefs may have more ﬂuid interactions with humans [17]. Underst”
Learning from Interventions: Human-robot interaction as both explicit and implicit feedback (2020, background): “We limit the scope of this work to a coarse model of human intervention that provides regret guarantees, leaving for future work the intriguing questions of potential differences between multiple experts, internal human state [30], trust [31], and cooperation [32, 33] in this setting.”
Shared autonomy via hindsight optimization for teleoperation and teaming (2017, background): “Similarly, Sadigh et al. (2016b) generate motions for an autonomous car using predictions of how other drivers will respond, enabling them to change the behavior of other users, and infer the internal user state (Sadigh et al., 2016a).”