• Corpus ID: 244463052

Visual Goal-Directed Meta-Learning with Contextual Planning Networks

@article{Rivera2021VisualGM,
  title={Visual Goal-Directed Meta-Learning with Contextual Planning Networks},
  author={Corban G. Rivera and David Handelman},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.09908}
}
The goal of meta-learning is to generalize to new tasks and goals as quickly as possible. Ideally, we would like approaches that generalize to new goals and tasks on the first attempt. Toward that end, we introduce contextual planning networks (CPN). Tasks are represented as goal images and used to condition the approach. We evaluate CPN along with several other approaches adapted for zero-shot goal-directed meta-learning. We evaluate these approaches across 24 distinct manipulation tasks using… 

Figures from this paper

References

SHOWING 1-10 OF 32 REFERENCES

Model-Based Planning in Discrete Action Spaces

This work introduces two discrete planning tasks inspired by existing question-answering datasets and shows that it is in fact possible to effectively perform planning via backprop in discrete action spaces using two simple yet principled modifications.

Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning

An open-source simulated benchmark for meta-reinforcement learning and multi-task learning consisting of 50 distinct robotic manipulation tasks is proposed to make it possible to develop algorithms that generalize to accelerate the acquisition of entirely new, held-out tasks.

Neural Task Programming: Learning to Generalize Across Hierarchical Tasks

A novel robot learning framework called Neural Task Programming (NTP), which bridges the idea of few-shot learning from demonstration and neural program induction, and achieves strong generalization across sequential tasks that exhibit hierarchal and compositional structures.

Hallucinative Topological Memory for Zero-Shot Visual Planning

This work proposes a simple VP method that plans directly in image space and significantly outperform the state-of-the-art VP methods, in terms of both plan interpretability and success rate when using the plan to guide a trajectory-following controller.

TACO: Learning Task Decomposition via Temporal Alignment for Control

This work proposes a weakly supervised, domain-agnostic approach based on task sketches, which include only the sequence of sub-tasks performed in each demonstration, and shows that this approach performs commensurately with fully supervised approaches, while requiring significantly less annotation effort.

Watch, Try, Learn: Meta-Learning from Demonstrations and Reward

This work proposes a method that can learn to learn from both demonstrations and trial-and-error experience with sparse reward feedback, and can scale to substantially broader distributions of tasks, as the demonstration reduces the burden of exploration.

Unsupervised Perceptual Rewards for Imitation Learning

This work presents a method that is able to identify key intermediate steps of a task from only a handful of demonstration sequences, and automatically identify the most discriminative features for identifying these steps.

Zero-Shot Visual Imitation

This workmitating expert demonstration is a powerful mechanism for learning to perform tasks from raw sensory observations by providing multiple demonstrations of a task at training time, and this generates data in the form of observation-action pairs from the agent's point of view.

The Predictron: End-To-End Learning and Planning

The predictron consists of a fully abstract model, represented by a Markov reward process, that can be rolled forward multiple "imagined" planning steps that accumulates internal rewards and values over multiple planning depths.

Deep spatial autoencoders for visuomotor learning

This work presents an approach that automates state-space construction by learning a state representation directly from camera images by using a deep spatial autoencoder to acquire a set of feature points that describe the environment for the current task, such as the positions of objects.