DeepAI AI Chat
Log In Sign Up

HyperDynamics: Meta-Learning Object and Agent Dynamics with Hypernetworks

by   Zhou Xian, et al.
Carnegie Mellon University

We propose HyperDynamics, a dynamics meta-learning framework that conditions on an agent's interactions with the environment and optionally its visual observations, and generates the parameters of neural dynamics models based on inferred properties of the dynamical system. Physical and visual properties of the environment that are not part of the low-dimensional state yet affect its temporal dynamics are inferred from the interaction history and visual observations, and are implicitly captured in the generated parameters. We test HyperDynamics on a set of object pushing and locomotion tasks. It outperforms existing dynamics models in the literature that adapt to environment variations by learning dynamics over high dimensional visual observations, capturing the interactions of the agent in recurrent state representations, or using gradient-based meta-optimization. We also show our method matches the performance of an ensemble of separately trained experts, while also being able to generalize well to unseen environment variations at test time. We attribute its good performance to the multiplicative interactions between the inferred system properties – captured in the generated parameters – and the low-dimensional state representation of the dynamical system.


page 8

page 14


Meta-World Conditional Neural Processes

We propose Meta-World Conditional Neural Processes (MW-CNP), a condition...

Meta-Learning with Latent Embedding Optimization

Gradient-based meta-learning techniques are both widely applicable and p...

Meta-learning generalizable dynamics from trajectories

We present the interpretable meta neural ordinary differential equation ...

MetaSDF: Meta-learning Signed Distance Functions

Neural implicit shape representations are an emerging paradigm that offe...

Contrastive Meta-Learning for Partially Observable Few-Shot Learning

Many contrastive and meta-learning approaches learn representations by i...

MERMAIDE: Learning to Align Learners using Model-Based Meta-Learning

We study how a principal can efficiently and effectively intervene on th...

A Neural-Symbolic Framework for Mental Simulation

We present a neural-symbolic framework for observing the environment and...