Search Results for author:
Found 22 papers, 3 papers with code
On Policy Reuse: An Expressive Language for Representing and Executing General Policies that Call Other Policies
Recently, a simple but powerful language for expressing and learning general policies and problem decompositions (sketches) has been introduced in terms of rules defined over a set of Boolean and numerical features.
Learning General Policies for Classical Planning Domains: Getting Beyond C$_2$
GNN-based approaches for learning general policies across planning domains are limited by the expressive power of $C_2$, namely; first-order logic with two variables and counting.
General Policies, Subgoal Structure, and Planning Width
It has been observed that many classical planning domains with atomic goals can be solved by means of a simple polynomial exploration procedure, called IW, that runs in time exponential in the problem width, which in these cases is bounded and small.
Language-Based Causal Representation Learning
We show that this is possible provided that the dynamics is learned over a suitable domain-independent first-order causal language that makes room for objects and relations that are not assumed to be known.
Learning Generalized Policies Without Supervision Using GNNs
We find for example that domains with general policies that require more expressive features can be solved with GNNs once the states are extended with suitable "derived atoms" encoding role compositions and transitive closures that do not fit into $C_{2}$.
Learning First-Order Symbolic Planning Representations That Are Grounded
In this work, we develop a new formulation for learning crisp first-order planning models that are grounded on parsed images, a step to combine the benefits of the two approaches.
Learning General Optimal Policies with Graph Neural Networks: Expressive Power, Transparency, and Limits
As predicted by the theory, it is observed that general optimal policies are obtained in domains where general optimal value functions can be defined with $C_2$ features but not in those requiring more expressive $C_3$ features.
Learning First-Order Representations for Planning from Black-Box States: New Results
For this, the learning problem is formulated as the search for a simplest first-order domain description D that along with information about instances I_i (number of objects and initial state) determine state space graphs G(P_i) that match the observed state graphs G_i where P_i = (D, I_i).
Flexible FOND Planning with Explicit Fairness Assumptions
The infinite trajectories that violate this condition are deemed as unfair, and the solutions are policies for which all the fair trajectories reach a goal state.
Learning General Policies from Small Examples Without Supervision
It has been recently shown that these policies can be computed in two steps: first, a suitable abstraction in the form of a qualitative numerical planning problem (QNP) is learned from sample plans, then the general policies are obtained from the learned QNP using a planner.
General Policies, Serializations, and Planning Width
We show that bounded width is a property of planning domains that admit optimal general policies in terms of features that are explicitly or implicitly represented in the domain encoding.
Qualitative Numeric Planning: Reductions and Complexity
This leads to a generate-and-test algorithm for solving QNPs where solutions to a FOND problem are generated one by one and tested for termination.
Higher-Dimensional Potential Heuristics for Optimal Classical Planning
Potential heuristics for state-space search are defined as weighted sums over simple state features.
Generalized Planning: Non-Deterministic Abstractions and Trajectory Constraints
Moreover, for a broad class of problems that involve integer variables that can be increased or decreased, trajectory constraints can be compiled away, reducing generalized planning to fully observable non-deterministic planning.
Action Selection for MDPs: Anytime AO* vs. UCT
In the presence of non-admissible heuristics, A* and other best-first algorithms can be converted into anytime optimal algorithms over OR graphs, by simply continuing the search after the first solution is found.
Factored Probabilistic Belief Tracking
The problem of belief tracking in the presence of stochastic actions and observations is pervasive and yet computationally intractable.
Causal Belief Decomposition for Planning with Sensing: Completeness Results and Practical Approximation
First, we introduce an alternative decomposition scheme and algorithm with the same time complexity but different completeness guarantees, whose space complexity is much smaller: exponential in the causal width of the problem that measures the number of state variables that are causally relevant to a given precondition, goal, or observable.
Learning First-Order Symbolic Representations for Planning from the Structure of the State Space
Solvers such as classical planners are very flexible and can deal with a variety of problem instances and goals but require first-order symbolic models.
Guarantees for Sound Abstractions for Generalized Planning (Extended Paper)
Recently it has been shown how to reduce the planning problem for generalized planning to the planning problem for a qualitative numerical problem; the latter being a reformulation that simultaneously captures all the instances in the collection.
Learning Features and Abstract Actions for Computing Generalized Plans
The actions in such plans, however, are not the actions in the instances themselves, which are not necessarily common to other instances, but abstract actions that are defined on a set of common features.
Features, Projections, and Representation Change for Generalized Planning
This is achieved by projecting the actions over the features, resulting in a common set of abstract actions which can be tested for soundness and completeness, and which can be used for generating general policies such as "if the gripper is empty, pick the clear block above x and place it on the table" that achieve the goal clear(x) in any Blocksworld instance.
Planning with Pixels in (Almost) Real Time
By using the same visual inputs, the planning results can be compared with those of humans and learning methods.
