Improves RL planning by proposing sub-goals hierarchically.
problem Sequential planning assumption in RL.
method Divide-and-Conquer Monte Carlo Tree Search (DC-MCTS).
result Improves navigation and control tasks.
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “task planning”
CoMPNetX uses neural networks to efficiently solve constrained motion planning problems.
problem Finding collision-free paths on constraint manifolds efficiently.
method Neural generator and discriminator with neural gradients-based projection operator.
result CoMPNetX finds path solutions with high success rates and lower computation times.
This paper introduces the QMDP-net, a neural network architecture for planning under partial observability. The QMDP-net combines the strengths of model-free learning and model-based planning. It is a recurrent policy network, but it represents a policy for a parameterized set of tasks by connecting a model with a plan…
This work improves AI's ability to solve physical tasks by optimizing world models in abstracted spaces.
problem Developing AI agents capable of solving diverse physical tasks and generalizing to new environments.
method Investigates and optimizes a family of joint-embedding predictive world models (JEPA-WMs) for efficient planning in abstracted spaces.
result Proposes a model that outperforms two established baselines in both navigation and manipulation tasks.
Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generationcs.LG
Video prediction models combined with planning algorithms have shown promise in enabling robots to learn to perform many vision-based tasks through only self-supervision, reaching novel goals in cluttered scenes with unseen objects. However, due to the compounding uncertainty in long horizon video prediction and poor s…
PackIt creates a virtual space for testing geometric planning skills.
problem Evaluating geometric planning abilities in virtual environments.
method Developed a virtual environment, PackIt, for geometric planning tasks.
result Demonstrated the effectiveness of various methods for geometric planning.
Planning has been very successful for control tasks with known environment dynamics. To leverage planning in unknown environments, the agent needs to learn the dynamics from interactions with the world. However, learning dynamics models that are accurate enough for planning has been a long-standing challenge, especiall…
Improved CEM for fast real-time planning in high-dimensional control tasks.
problem Sampling inefficiency of CEM in real-time planning.
method Novel additions to CEM including temporally-correlated actions and memory.
result 2.7-22x less samples and 1.2-10x performance increase.
Agents struggle with solving tasks in new environments, but new models improve performance.
problem Rapid task-solving in novel environments.
method Developed EPNs to enable deep RL agents to plan over gathered knowledge.
result EPNs enable deep RL agents to excel at RTS, outperforming baselines by factors of 2-3.
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…
Plan2Explore learns new tasks efficiently through self-supervised planning.
problem Challenges in reinforcement learning, especially task-specific learning and sample efficiency.
method Self-supervised exploration and fast adaptation to new tasks through efficient planning.
result Plan2Explore outperforms prior methods in learning new tasks without supervision.
Integrates skills and world models for efficient task solving and transfer.
problem Quickly solve new tasks in complex environments using reusable knowledge.
method Leverages partial amortization for fast adaptation and online skill planning.
result Improved sample efficiency in single tasks and transfer between tasks.
Deep network predicts action sequences for complex tasks from a scene image.
problem Scalable task and motion planning from initial scene images.
method Deep convolutional recurrent neural network that predicts action sequences.
result Predicts promising action sequences, reducing motion planning problems.
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…
In reinforcement learning, the discount factor controls the agent's effective planning horizon. Traditionally, this parameter was considered part of the MDP; however, as deep reinforcement learning algorithms tend to become unstable when the effective planning horizon is long, recent works refer to as a hyper-p…
Agents compose pre-trained policies for complex tasks, improving zero-shot performance.
problem Challenges in long-horizon predictions and estimating visitation distributions induced by policy sequences.
method Learn predictive jumpy world models of multi-step dynamics, enhancing predictions with a consistency objective.
result Compositional planning with jumpy world models yields, on average, a 200% relative improvement over primitive actions on long-horizon tasks.
We investigate the integration of a planning mechanism into sequence-to-sequence models using attention. We develop a model which can plan ahead in the future when it computes its alignments between input and output sequences, constructing a matrix of proposed future alignments and a commitment vector that governs whet…
This work tackles long-term visual planning by goal-conditioned hierarchical predictors.
problem Current learning approaches fail on long-horizon tasks due to lack of goal information and coarse-to-fine planning.
method Formulate goal-conditioned predictors (GCPs) and hierarchical models to predict trajectories between observations.
result GCPs enable effective long-term planning with much longer horizons than before.
Certified guidance ensures generative models always meet planning objectives.
problem Lack of guarantees for generated outputs satisfying planning objectives.
method Certified guidance modifies generative models without retraining, ensuring outputs meet STL specifications.
result Certified guidance produces models that are always correct, unlike non-certified methods.
EDGI improves sample efficiency and generalization in tasks with spatial and temporal symmetries.
problem Sample inefficiency and poor generalization in tasks with geometric symmetries.
method Equivariant Diffuser framework, SE(3)xZxSn-equivariant diffusion model.
result EDGI is more sample efficient and generalizes better than non-equivariant models.
Algorithm learns multiple tasks with minimal planning, achieving near-optimal performance.
problem Learning multiple tasks efficiently in a reinforcement learning setting.
method UCB Lifelong Value Distillation (UCBlvd) algorithm with structural assumption for shared exploration.
result Sublinear regret bound of with planning calls.
Learning and inference movement is a very challenging problem due to its high dimensionality and dependency to varied environments or tasks. In this paper, we propose an effective probabilistic method for learning and inference of basic movements. The motion planning problem is formulated as learning on a directed grap…
Paper tackles continual reinforcement learning by forgetting, proposing a planning method with online world models.
problem Catastrophic forgetting in reinforcement learning when learning new tasks.
method Planning with an online world model using model predictive control.
result The proposed FTL Online Agent (OA) learns new tasks without forgetting old skills.
Develops a method to plan exploration that learns strong policies with fewer samples.
problem Lack of efficient exploration in reinforcement learning for real-world tasks.
method Plans an action sequence that maximizes information gain about the optimal trajectory.
result 2x fewer samples than exploration baselines and 200x fewer than model-free methods.
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 …
We introduce the value iteration network (VIN): a fully differentiable neural network with a `planning module' embedded within. VINs can learn to plan, and are suitable for predicting outcomes that involve planning-based reasoning, such as policies for reinforcement learning. Key to our approach is a novel differentiab…
We demonstrate the use of conditional autoregressive generative models (van den Oord et al., 2016a) over a discrete latent space (van den Oord et al., 2017b) for forward planning with MCTS. In order to test this method, we introduce a new environment featuring varying difficulty levels, along with moving goals and obst…
We present HiDe, a novel hierarchical reinforcement learning architecture that successfully solves long horizon control tasks and generalizes to unseen test scenarios. Functional decomposition between planning and low-level control is achieved by explicitly separating the state-action spaces across the hierarchy, which…
Planning methods can solve temporally extended sequential decision making problems by composing simple behaviors. However, planning requires suitable abstractions for the states and transitions, which typically need to be designed by hand. In contrast, model-free reinforcement learning (RL) can acquire behaviors from l…
Plan2Vec learns image representations without labels, improving control tasks.
problem Learning image representations without labeled data.
method Constructs a weighted graph using near-neighbor distances and extrapolates to global embedding.
result Plan2Vec achieves accurate long-term value estimates in control tasks with reduced computational and memory costs.
The objective of this work is to augment the basic abilities of a robot by learning to use new sensorimotor primitives to enable the solution of complex long-horizon problems. Solving long-horizon problems in complex domains requires flexible generative planning that can combine primitive abilities in novel combination…
Manipulation tasks such as preparing a meal or assembling furniture remain highly challenging for robotics and vision. Traditional task and motion planning (TAMP) methods can solve complex tasks but require full state observability and are not adapted to dynamic scene changes. Recent learning methods can operate direct…
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…
We introduce a new reinforcement learning approach combining a planning quasi-metric (PQM) that estimates the number of steps required to go from any state to another, with task-specific "aimers" that compute a target state to reach a given goal. This decomposition allows the sharing across tasks of a task-agnostic mod…
Olympus benchmarks optimization algorithms for noisy experiments.
problem Benchmarking optimization algorithms on realistic experimental scenarios is challenging.
method Introduces Olympus, a software package for benchmarking optimization algorithms on synthetic experiments.
result Mitigates barriers in benchmarking optimization algorithms on realistic experimental scenarios.
Autonomous agents situated in real-world environments must be able to master large repertoires of skills. While a single short skill can be learned quickly, it would be impractical to learn every task independently. Instead, the agent should share knowledge across behaviors such that each task can be learned efficientl…
We present a framework for autonomously learning a portable representation that describes a collection of low-level continuous environments. We show that these abstract representations can be learned in a task-independent egocentric space specific to the agent that, when grounded with problem-specific information, are …
We introduce an algorithm for model-based hierarchical reinforcement learning to acquire self-contained transition and reward models suitable for probabilistic planning at multiple levels of abstraction. We call this framework Planning with Abstract Learned Models (PALM). By representing subtasks symbolically using a n…
Value-based methods constitute a fundamental methodology in planning and deep reinforcement learning (RL). In this paper, we propose to exploit the underlying structures of the state-action value function, i.e., Q function, for both planning and deep RL. In particular, if the underlying system dynamics lead to some glo…
Some of the most important tasks take place in environments which lack cheap and perfect simulators, thus hampering the application of model-free reinforcement learning (RL). While model-based RL aims to learn a dynamics model, in a more general case the learner does not know a priori what the action space is. Here we …
This work clarifies the role of inference types in planning.
problem Lack of consistency in using inference types for planning.
method Variational framework and loopy belief propagation.
result All inference types correspond to different weights in variational problems.
Framework learns useful subgoals from demonstrations and instructions.
problem Efficient long-term planning for novel goals.
method Rational subgoals (RSGs) learned from demonstrations and task descriptions.
result Improves performance-time efficiency of planning algorithms.
PlanGAN uses GANs to plan efficient trajectories for multi-goal tasks in sparse reward environments.
problem Learning with sparse rewards in multi-goal environments.
method PlanGAN combines GANs to generate trajectories leading to specified goals, then combines these into a planning algorithm.
result PlanGAN achieves comparable performance to model-free RL but is 4-8 times more sample efficient.
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…
NOT learns optimal transport plans, kernel costs improve performance.
problem NOT algorithm learns non-optimal plans with weak quadratic costs.
method Introduced kernel weak quadratic costs to improve NOT's performance.
result Kernel costs provide improved theoretical and practical guarantees.
New method optimizes experiments under constraints.
problem Adapting BED to dynamic constraints in real-world tasks.
method Offline pre-training of an amortized policy and posterior network with online multi-step lookahead planning.
result Significantly more informative design sequences than existing methods.
Sliced-regularized OT improves transport plan accuracy.
problem Optimal transport (OT) approximation accuracy.
method Sliced-regularized optimal transport (SROT) formulation.
result SROT yields more accurate approximations of exact OT than entropic OT.
Predicting outcomes and planning interactions with the physical world are long-standing goals for machine learning. A variety of such tasks involves continuous physical systems, which can be described by partial differential equations (PDEs) with many degrees of freedom. Existing methods that aim to control the dynamic…