Deep generative models for offline policy learning: Tutorial, survey, and perspectives on future directions
Deep generative models (DGMs) have demonstrated great success across various domains,
particularly in generating texts, images, and videos using models trained from offline data …
particularly in generating texts, images, and videos using models trained from offline data …
Learning sampling distributions for model predictive control
Sampling-based methods have become a cornerstone of contemporary approaches to
Model Predictive Control (MPC), as they make no restrictions on the differentiability of the …
Model Predictive Control (MPC), as they make no restrictions on the differentiability of the …
Dual rl: Unification and new methods for reinforcement and imitation learning
The goal of reinforcement learning (RL) is to find a policy that maximizes the expected
cumulative return. It has been shown that this objective can be represented as an …
cumulative return. It has been shown that this objective can be represented as an …
Flomo: Tractable motion prediction with normalizing flows
C Schöller, A Knoll - … on Intelligent Robots and Systems (IROS), 2021 - ieeexplore.ieee.org
The future motion of traffic participants is inherently uncertain. To plan safely, therefore, an
autonomous agent must take into account multiple possible trajectory outcomes and …
autonomous agent must take into account multiple possible trajectory outcomes and …
[PDF][PDF] Diverse sampling for normalizing flow based trajectory forecasting
Normalizing flows have recently emerged as an attractive model for autonomous vehicle
trajectory forecasting. However, a key drawback is that iid samples from flow models often …
trajectory forecasting. However, a key drawback is that iid samples from flow models often …
Individual survival curves with conditional normalizing flows
Survival analysis, or time-to-event modelling, is a classical statistical problem that has
garnered a lot of interest for its practical use in epidemiology, demographics or actuarial …
garnered a lot of interest for its practical use in epidemiology, demographics or actuarial …
Sampling for Model Predictive Trajectory Planning in Autonomous Driving using Normalizing Flows
G Rabenstein, L Ullrich, K Graichen - arXiv preprint arXiv:2404.09657, 2024 - arxiv.org
Alongside optimization-based planners, sampling-based approaches are often used in
trajectory planning for autonomous driving due to their simplicity. Model predictive path …
trajectory planning for autonomous driving due to their simplicity. Model predictive path …
Robust and Probabilistic Motion Prediction for Intelligent Infrastructure Systems
C Schöller - 2022 - mediatum.ub.tum.de
Autonomous driving promises various benefits to its users, such as improved comfort, more
ecological transportation, and–most importantly–higher safety than manual driving. One …
ecological transportation, and–most importantly–higher safety than manual driving. One …
[图书][B] Learning Novel Strategies for Model Predictive Control by Leveraging Experience
JI Sacks - 2023 - search.proquest.com
A major challenge in robotics is to design robust policies which enable complex and agile
behaviors in the real world. On one end of the spectrum, we have model-free reinforcement …
behaviors in the real world. On one end of the spectrum, we have model-free reinforcement …
Generalized flow-based variational autoencoder networks for anomaly detection in multivariate time series
R Shah - 2021 - ideals.illinois.edu
Time series are widely used in applications such as finance, robotics, telecommunications,
astronomy, and many more. Detecting anomalies like robotic arm failures or server attacks is …
astronomy, and many more. Detecting anomalies like robotic arm failures or server attacks is …