• Publications
  • Influence
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
TLDR
This paper proposes a new algorithm called probabilistic ensembles with trajectory sampling (PETS) that combines uncertainty-aware deep network dynamics models with sampling-based uncertainty propagation, which matches the asymptotic performance of model-free algorithms on several challenging benchmark tasks, while requiring significantly fewer samples. Expand
Manifold Gaussian Processes for regression
TLDR
Manifold Gaussian Processes is a novel supervised method that jointly learns a transformation of the data into a feature space and a GP regression from the feature space to observed space, which allows to learn data representations, which are useful for the overall regression task. Expand
Learning Invariant Representations for Reinforcement Learning without Reconstruction
TLDR
This work studies how representation learning can accelerate reinforcement learning from rich observations, such as images, without relying either on domain knowledge or pixel-reconstruction, and proposes a method to learn robust latent representations which encode only the task-relevant information from observations. Expand
Bayesian optimization for learning gaits under uncertainty
TLDR
Bayesian optimization, a model-based approach to black-box optimization under uncertainty, is evaluated on both simulated problems and real robots, demonstrating that Bayesian optimization is particularly suited for robotic applications, where it is crucial to find a good set of gait parameters in a small number of experiments. Expand
The Feeling of Success: Does Touch Sensing Help Predict Grasp Outcomes?
TLDR
This work investigated the question of whether touch sensing aids in predicting grasp outcomes within a multimodal sensing framework that combines vision and touch, and evaluated visuo-tactile deep neural network models to directly predict grasp outcomes from either modality individually, and from both modalities together. Expand
An experimental comparison of Bayesian optimization for bipedal locomotion
TLDR
This paper experimentally evaluated some common methods for automatic gait optimization in bipedal locomotion, and analyzed Bayesian optimization in different configurations, including various acquisition functions. Expand
Adversarial Continual Learning
TLDR
This work proposes a novel hybrid continual learning framework that learns a disjoint representation for task-invariant and task-specific features required to solve a sequence of tasks. Expand
More Than a Feeling: Learning to Grasp and Regrasp Using Vision and Touch
TLDR
An end-to-end action-conditional model that learns regrasping policies from raw visuo-tactile data and outperforms a variety of baselines at estimating grasp adjustment outcomes, selecting efficient grasp adjustments for quick grasping, and reducing the amount of force applied at the fingers, while maintaining competitive performance. Expand
Learning Deep Belief Networks from Non-stationary Streams
TLDR
This paper proposes a proof-of-concept method, called Adaptive Deep Belief Networks, of how deep learning can be generalized to learn online from changing streams of data by exploiting the generative properties of the model to incrementally re-train the Deep Belief Network whenever new data are collected. Expand
Low-Level Control of a Quadrotor With Deep Model-Based Reinforcement Learning
TLDR
This is the first use of MBRL for controlled hover of a quadrotor using only on-board sensors, direct motor input signals, and no initial dynamics knowledge, and the controller leverages rapid simulation of a neural network forward dynamics model on a graphic processing unit enabled base station. Expand
...
1
2
3
4
5
...