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.
This work defines a complexity measure for BAMDP planning and introduces state abstraction for more efficient approximate planning.
In this paper we study a notion of topological complexity for the motion planning problem. The topological complexity is a number which measures discontinuity of the process of motion planning in the configuration space X. More precisely, it is the minimal number k such that there are k different motion planning rules,…
Survey on manifold complexities and motion planning in robotics.
Study motion planning for points avoiding obstacles in a plane.
CoMPNetX uses neural networks to efficiently solve constrained motion planning problems.
Paper analyzes neural network complexity for planning problems.
New TC variant dTC better fits motion planning for some systems.
Deep network predicts action sequences for complex tasks from a scene image.
We consider the problem of robot motion planning in an oriented Riemannian manifold as a topological motion planning problem in its oriented frame bundle. For this purpose, we study the topological complexity of oriented frame bundles, derive an upper bound for this invariant and certain lower bounds from cup length co…
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 …
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 algorithm reduces sample complexity for planning in MDPs.
Proposes a method for model-based RL in complex environments without perfect simulators.
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 consider the problem of online planning in a Markov Decision Process when given only access to a generative model, restricted to open-loop policies - i.e. sequences of actions - and under budget constraint. In this setting, the Open-Loop Optimistic Planning (OLOP) algorithm enjoys good theoretical guarantees but is …
Vanguard uses AI to create personalized financial plans.
HiDe learns hierarchical control for complex tasks by separating planning and control.
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…
Study geodesic complexity in homogeneous Riemannian manifolds.
Constructing agents with planning capabilities has long been one of the main challenges in the pursuit of artificial intelligence. Tree-based planning methods have enjoyed huge success in challenging domains, such as chess and Go, where a perfect simulator is available. However, in real-world problems the dynamics gove…
Let X be a subcomplex of the standard CW-decomposition of the n-dimensional torus. We exhibit an explicit optimal motion planning algorithm for X. This construction is used to calculate the topological complexity of complements of general position arrangements and Eilenberg-Mac Lane spaces associated to right-angled Ar…
Improved API to achieve optimal error bound and query complexity in local planning.
State-of-the-art efficient model-based Reinforcement Learning (RL) algorithms typically act by iteratively solving empirical models, i.e., by performing \emph{full-planning} on Markov Decision Processes (MDPs) built by the gathered experience. In this paper, we focus on model-based RL in the finite-state finite-horizon…
The topological complexity TC(X) is a numerical homotopy invariant of a topological space X which is motivated by robotics and is similar in spirit to the classical Lusternik-Schnirelmann category of X. Given a mechanical system with configuration space X, the invariant TC(X) measures the complexity of all possible mot…
Agents compose pre-trained policies for complex tasks, improving zero-shot performance.
This paper uses probability tensors for efficient path planning in complex scenarios.
This paper studies when particle filtering is efficient for planning in partially observed systems.
Study experiment planning with function approximation in contextual bandit problems.
The field of reinforcement learning (RL) is facing increasingly challenging domains with combinatorial complexity. For an RL agent to address these challenges, it is essential that it can plan effectively. Prior work has typically utilized an explicit model of the environment, combined with a specific planning algorith…
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…
The paper tackles model-based RL's inaccuracy issue by dynamically adjusting planning horizons.
We prove that the topological complexity of (a motion planning algorithm on) the complement of generic complex essential hyperplane arrangement of hyperplanes in an -dimensional linear space is min.
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…
The explore{exploit dilemma is one of the central challenges in Reinforcement Learning (RL). Bayesian RL solves the dilemma by providing the agent with information in the form of a prior distribution over environments; however, full Bayesian planning is intractable. Planning with the mean MDP is a common myopic approxi…
A major difficulty of solving continuous POMDPs is to infer the multi-modal distribution of the unobserved true states and to make the planning algorithm dependent on the perceived uncertainty. We cast POMDP filtering and planning problems as two closely related Sequential Monte Carlo (SMC) processes, one over the real…
Plan2Vec learns image representations without labels, improving control tasks.
Unified framework for planning under uncertainty using variational inference.
Efficient algorithms for planning in cooperative multi-agent reinforcement learning with combinatorial action spaces.
Deep RL drone trained to compete against classical path planning in drone racing.
TensorPlan shows an exponential lower bound for planning in MDPs with linearly realizable value functions.
Model-based reinforcement learning (MBRL) with model-predictive control or online planning has shown great potential for locomotion control tasks in terms of both sample efficiency and asymptotic performance. Despite their initial successes, the existing planning methods search from candidate sequences randomly generat…
New metric invariant GC connects to TC, with applications in robotics.
We study an elementary problem of topological robotics: rotation of a line, which is fixed by a revolving joint at a base point: one wants to bring the line from its initial position to a final position by a continuous motion in the space. The final goal is to construct an algorithm which will perform this task once th…
PAC-MCTS addresses biased search in LLM-guided planning by dynamically pruning.
New algorithm for RL with horizon-free reward-free exploration for linear MDPs.
Novel method controls complex physical systems over long time frames.
Deep learning improves trip prediction accuracy in transportation planning.
Paper proposes incorporating road rules as a loss function for better motion planning.