Learning deep dynamical models from image pixels

By Niklas Wahlstrom, Thomas Bo Schön, Marc Peter Deisenroth. 2014

67 citations recorded by the source corpus.

Modeling dynamical systems is important in many disciplines, such as control, robotics, or neurotechnology. Commonly the state of these systems is not directly observed, but only available through ...

Subfields: Artificial Intelligence, Biophysics, Statistical and Nonlinear Physics, Generative Models

How researchers see this paper

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

In Artificial Intelligence

In Biophysics

In Statistical and Nonlinear Physics

In Generative Models

Citations in the corpus

From Pixels to Torques: Policy Learning with Deep Dynamical Models (2015, methodology): “mensional embedded time series. The property of fast predictions will be exploited by the online feedback control strategy presented in the following. More details on the proposed model are given in (Wahlstrom et al., 2015).¨ 3. Learning Closed-Loop Policies from We use the DDM for learning a closed”
Recurrent World Models Facilitate Policy Evolution (2018, background): “Recent work along these lines [99, 100] was able to train controllers using the bottleneck hidden layer of an autoencoder as low-dimensional feature vectors to control a pendulum from pixel inputs.”
Deep Reinforcement Learning for Autonomous Driving: A Survey (2020, background): “Model-based deep RL algorithms have been proposed for learning models and policies directly from raw pixel inputs [91], [92].”
Manifold Gaussian Processes for regression (2014, background): “feature learning can be insufficient as the learned data representation H might not suit the overall regression task F (Wahlström et al., 2015): Unsupervised and supervised learning optimize different objectives, which do not necessarily match, e.”
A Brief Survey of Deep Reinforcement Learning (2017, positive): “Based on the deep dynamical model [131], where high-dimensional observations are embedded into a lower-dimensional space using autoencoders, several model-based DRL algorithms have been proposed for learning models and policies from pixel information [81, 137, 132].”
Deep predictive policy training using reinforcement learning (2017, methodology): “More recently, MPC is accompanied by general function approximators, such as Gaussian processes [14], [24] or artificial neural networks [10] to learn the dynamic model directly from interaction data.”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “Semantic mapping is much more than a categorization problem: The semantic concepts are evolving to more specialized information such as affordances and actionability 4 of the entities in the map and the possible interactions among different active agents in the environment.”
Data-Efficient Learning of Feedback Policies from Image Pixels using Deep Dynamical Models (2015, methodology): “r images to similar features, which led to zig-zagging around in feature space (instead of following a smooth manifold as in Figure 4), making the model learning part in feature space incredibly hard [17]. We modeled and controlled equally complex models with E2C [16], but at the same time our DDM r”
Interactive Perception: Leveraging Action in Perception and Perception in Action (2016, methodology): “, linear mapping with task-specific regularizers [71] or Convolutional Neural Networks [73], [74].”