Efficient high dimensional maximum entropy modeling via symmetric partition functions

By Paul Vernaza, Drew Bagnell. 2012

23 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

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

In Probabilistic & Bayesian Methods

Citations in the corpus

Learning Trajectory Preferences for Manipulators via Iterative Improvement (2013, neutral): “Previously both approaches [3, 4, 6, 36] have been studied.”
Socially compliant mobile robot navigation via inverse reinforcement learning (2016, methodology): “Whereas Demyen and Buro (2006) assumed a polygonal representation of the environment, we allow arbitrary obstacles on grid maps, which enables us to incorporate online real-world sensor data.”
Learning preferences for manipulation tasks from online coactive feedback (2015, background): “Two effective ways of solving the former problem are either discretizing the state space (Alterovitz et al., 2007; Bhattacharya et al., 2011; Vernaza and Bagnell, 2012) or directly sampling trajectories from the continuous space (Berg et al.”
A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress (2018, unknown): “Other approaches for likelihood approximation in a high-dimensional state space include the use of importance sampling by reirl, the down-scaling the state space using low-dimensional features [62], and the assumption by pi-irl that demonstrations are locally optimal.”
Approximate MaxEnt Inverse Optimal Control and Its Application for Mental Simulation of Human Interactions (2015, background): “The second insight, which is also used in some previous work such as (Vernaza and Bagnell 2012), is that one can estimate an expectation by Monte Carlo sampling and the error behavior would beO( 1√ N ) (forN independent trajectories), Algorithm 1 – Backward pass D(t)m = {(Xi, Ai, Rti, X ′i)}mi=1,…”