arXiv research

Improves RL planning by proposing sub-goals hierarchically.

A key challenge in complex visuomotor control is learning abstract representations that are effective for specifying goals, planning, and generalization. To this end, we introduce universal planning networks (UPN). UPNs embed differentiable planning within a goal-directed policy. This planning computation unrolls a for…

Imitation Learning (IL) is an appealing approach to learn desirable autonomous behavior. However, directing IL to achieve arbitrary goals is difficult. In contrast, planning-based algorithms use dynamics models and reward functions to achieve goals. Yet, reward functions that evoke desirable behavior are often difficul…

In recent years, deep generative models have been shown to 'imagine' convincing high-dimensional observations such as images, audio, and even video, learning directly from raw data. In this work, we ask how to imagine goal-directed visual plans -- a plausible sequence of observations that transition a dynamical system …

Many AI problems, in robotics and other domains, are goal-directed, essentially seeking a trajectory leading to some goal state. In such problems, the way we choose to represent a trajectory underlies algorithms for trajectory prediction and optimization. Interestingly, most all prior work in imitation and reinforcemen…

Goal-directed manipulation of representations is a key element of human flexible behaviour, while consciousness is often related to several aspects of higher-order cognition and human flexibility. Currently these two phenomena are only partially integrated (e.g., see Neurorepresentationalism) and this (a) limits our un…

Agents need world models to generalize multi-step tasks.

Paper proposes a method to make learned models focus on task-relevant information.

New RL approach infers optimal policies via variational inference.

Extended Predictable Feature Analysis (PFAx) [Richthofer and Wiskott, 2017] is an extension of PFA [Richthofer and Wiskott, 2015] that allows generating a goal-directed control signal of an agent whose dynamics has previously been learned during a training phase in an unsupervised manner. PFAx hardly requires assumptio…

New method learns optimal environment and goal difficulty for reinforcement learning.

NIPA aims to translate brain learning mechanisms into scalable Bayesian inference.

Paper tackles goal-directed generation of discrete structures using conditional generative models.

An internal model of the own body can be assumed a fundamental and evolutionary-early representation as it is present throughout the animal kingdom. Such functional models are, on the one hand, required in motor control, for example solving the inverse kinematic or dynamic task in goal-directed movements or a forward t…

Causal reasoning has been an indispensable capability for humans and other intelligent animals to interact with the physical world. In this work, we propose to endow an artificial agent with the capability of causal reasoning for completing goal-directed tasks. We develop learning-based approaches to inducing causal kn…

We propose a hybrid approach aimed at improving the sample efficiency in goal-directed reinforcement learning. We do this via a two-step mechanism where firstly, we approximate a model from Model-Free reinforcement learning. Then, we leverage this approximate model along with a notion of reachability using Mean First P…

Chance-constrained ActInf allows for small violations of constraints to drive goal-directed behavior.

Generating novel graph structures that optimize given objectives while obeying some given underlying rules is fundamental for chemistry, biology and social science research. This is especially important in the task of molecular graph generation, whose goal is to discover novel molecules with desired properties such as …

Consider mutli-goal tasks that involve static environments and dynamic goals. Examples of such tasks, such as goal-directed navigation and pick-and-place in robotics, abound. Two types of Reinforcement Learning (RL) algorithms are used for such tasks: model-free or model-based. Each of these approaches has limitations.…

Sparse reward problems are one of the biggest challenges in Reinforcement Learning. Goal-directed tasks are one such sparse reward problems where a reward signal is received only when the goal is reached. One promising way to train an agent to perform goal-directed tasks is to use Hindsight Learning approaches. In thes…

Universal AI seeks high-optionality states through empowerment and curiosity.

Graph clustering is a fundamental task which discovers communities or groups in networks. Recent studies have mostly focused on developing deep learning approaches to learn a compact graph embedding, upon which classic clustering methods like k-means or spectral clustering algorithms are applied. These two-step framewo…

Animals execute goal-directed behaviours despite the limited range and scope of their sensors. To cope, they explore environments and store memories maintaining estimates of important information that is not presently available. Recently, progress has been made with artificial intelligence (AI) agents that learn to per…

New method designs fairer transport plans with uncertainty.

We aim to reduce the burden of programming and deploying autonomous systems to work in concert with people in time-critical domains, such as military field operations and disaster response. Deployment plans for these operations are frequently negotiated on-the-fly by teams of human planners. A human operator then trans…

Study integrates reliability constraints into generation planning models.

New approach improves black-box planning efficiency by discovering focused macros.

Study motion planning for points avoiding obstacles in a plane.

Selective planning with imperfect models reduces harmful effects of model inadequacy.

CoMPNetX uses neural networks to efficiently solve constrained motion planning problems.

This article asks how planning scholarship may effectively gain impact in planning practice through media exposure. In liberal democracies the public sphere is dominated by mass media. Therefore, working with such media is a prerequisite for effective public impact of planning research. Using the example of megaproject…

We introduce Dynamic Planning Networks (DPN), a novel architecture for deep reinforcement learning, that combines model-based and model-free aspects for online planning. Our architecture learns to dynamically construct plans using a learned state-transition model by selecting and traversing between simulated states and…

This paper presents a unifying framework for reinforcement learning and planning.

A planning approach learns skills from interactions, balancing exploration and exploitation.

Survey of integrating planning and learning in model-based reinforcement learning.

Fast and efficient motion planning algorithms are crucial for many state-of-the-art robotics applications such as self-driving cars. Existing motion planning methods become ineffective as their computational complexity increases exponentially with the dimensionality of the motion planning problem. To address this issue…

New approach for obstacle avoidance in robotics using learned representations.

Reinforcement learning and symbolic planning have both been used to build intelligent autonomous agents. Reinforcement learning relies on learning from interactions with real world, which often requires an unfeasibly large amount of experience. Symbolic planning relies on manually crafted symbolic knowledge, which may …

This work clarifies the role of inference types in planning.

This work tackles long-term visual planning by goal-conditioned hierarchical predictors.

Knowledge-based planning (KBP) is an automated approach to radiation therapy treatment planning that involves predicting desirable treatment plans before they are then corrected to deliverable ones. We propose a generative adversarial network (GAN) approach for predicting desirable 3D dose distributions that eschews th…

We designed a grid world task to study human planning and re-planning behavior in an unknown stochastic environment. In our grid world, participants were asked to travel from a random starting point to a random goal position while maximizing their reward. Because they were not familiar with the environment, they needed…

Efficient driving in urban traffic scenarios requires foresight. The observation of other traffic participants and the inference of their possible next actions depending on the own action is considered cooperative prediction and planning. Humans are well equipped with the capability to predict the actions of multiple i…

This work defines a complexity measure for BAMDP planning and introduces state abstraction for more efficient approximate planning.

Information planning enables faster learning with fewer training examples. It is particularly applicable when training examples are costly to obtain. This work examines the advantages of information planning for text data by focusing on three supervised models: Naive Bayes, supervised LDA and deep neural networks. We s…

The computational costs of inference and planning have confined Bayesian model-based reinforcement learning to one of two dismal fates: powerful Bayes-adaptive planning but only for simplistic models, or powerful, Bayesian non-parametric models but using simple, myopic planning strategies such as Thompson sampling. We …

Conventional wisdom holds that model-based planning is a powerful approach to sequential decision-making. It is often very challenging in practice, however, because while a model can be used to evaluate a plan, it does not prescribe how to construct a plan. Here we introduce the "Imagination-based Planner", the first m…

CoTj improves diffusion model quality and stability via graph planning.