Related papers: Neural Modular Control for Embodied Question Answe…

Policy learning for partially observed control tasks requires policies that can remember salient information from past observations. In this paper, we present a method for learning policies with internal memory for high-dimensional,…

Model-based planners and controllers are commonly used to solve complex manipulation problems as they can efficiently optimize diverse objectives and generalize to long horizon tasks. However, they often fail during deployment due to noisy…

We propose an active learning architecture for robots, capable of organizing its learning process to achieve a field of complex tasks by learning sequences of motor policies, called Intrinsically Motivated Procedure Babbling (IM-PB). The…

We study the problem of programmatic reinforcement learning, in which policies are represented as short programs in a symbolic language. Programmatic policies can be more interpretable, generalizable, and amenable to formal verification…

Most reinforcement-learning (RL) controllers used in continuous control are architecturally centralized: observations are compressed into a single latent state from which both value estimates and actions are produced. Biological control…

Goal-conditioned reinforcement learning endows an agent with a large variety of skills, but it often struggles to solve tasks that require more temporally extended reasoning. In this work, we propose to incorporate imagined subgoals into…

Many potential applications of reinforcement learning in the real world involve interacting with other agents whose numbers vary over time. We propose new neural policy architectures for these multi-agent problems. In contrast to other…

This work introduces a growable and modular neural network architecture that naturally avoids catastrophic forgetting and interference in continual reinforcement learning. The structure of each module allows the selective combination of…

Control policy learning for modular robot locomotion has previously been limited to proprioceptive feedback and flat terrain. This paper develops policies for modular systems with vision traversing more challenging environments. These…

End-to-end learning of robot control policies, structured as neural networks, has emerged as a promising approach to robotic manipulation. To complete many common tasks, relevant objects are required to pass in and out of a robot's field of…

This paper presents a novel framework for automatic learning of complex strategies in human decision making. The task that we are interested in is to better facilitate long term planning for complex, multi-step events. We observe temporal…

Biological intelligence can learn to solve many diverse tasks in a data efficient manner by re-using basic knowledge and skills from one task to another. Furthermore, many of such skills are acquired without explicit supervision in an…

Learning to navigate in complex environments with dynamic elements is an important milestone in developing AI agents. In this work we formulate the navigation question as a reinforcement learning problem and show that data efficiency and…

Rather than learning new control policies for each new task, it is possible, when tasks share some structure, to compose a "meta-policy" from previously learned policies. This paper reports results from experiments using Deep Reinforcement…

For safe and efficient planning and control in autonomous driving, we need a driving policy which can achieve desirable driving quality in long-term horizon with guaranteed safety and feasibility. Optimization-based approaches, such as…

Conventionally, memory in end-to-end robotic learning involves inputting a sequence of past observations into the learned policy. However, in complex multi-stage real-world tasks, the robot's memory must represent past events at multiple…

Neural Module Networks (NMN) are a compelling method for visual question answering, enabling the translation of a question into a program consisting of a series of reasoning sub-tasks that are sequentially executed on the image to produce…

Embodied Question Answering (EQA) is a recently proposed task, where an agent is placed in a rich 3D environment and must act based solely on its egocentric input to answer a given question. The desired outcome is that the agent learns to…

In hierarchical reinforcement learning a major challenge is determining appropriate low-level policies. We propose an unsupervised learning scheme, based on asymmetric self-play from Sukhbaatar et al. (2018), that automatically learns a…

Deep reinforcement learning for high dimensional, hierarchical control tasks usually requires the use of complex neural networks as functional approximators, which can lead to inefficiency, instability and even divergence in the training…