Planning as Search: A Quantitative Approach

By Richard E. Korf. 1987

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

Citations in the corpus

Pre-Trained Language Models for Interactive Decision-Making (2022, methodology): “We build on the original environment to create the VirtualHomeImitation Learning dataset using regression planning (Korf, 1987), which features more challenging tasks and more diverse scenes and objects.”
RoboCup: The Robot World Cup Initiative (1997, background): “Also, Korf addressed it theoretically in [Korf 87].”
Feature-based methods for large scale dynamic programming (1995, background): “In artificial intelligence, this issue is usually addressed by using heuristic position evaluation functions; chess playing programs are a prime example (Korf, 1987).”
Heuristic shortest path algorithms for transportation applications: State of the art (2006, background): “The concept was further explored by Sacerdoti [48], Korf [49] and many other researchers [50–55].”
Regression Planning Networks (2019, methodology): “Similar to forward planning, backward planning in symbolic space (formally known as regression planning [9, 10] or pre-image backchaining [11–13]) also relies on a planning domain to expand the search space starting from the final goal until the current state is reached.”
Building Cooperative Embodied Agents Modularly with Large Language Models (2023, methodology): “5.2 B ASELINES MCTS-based Hierarchical Planner(MHP) is adopted from the strongest baseline in the original Watch-And-Help Challenge, which is a Hierarchical Planner with a high-level planner based on MCTS and a low-level planner based on regression planning (Korf, 1987).”
Automated Planning and Acting (2016, background): “Korf [352] introduced distinctions among problems with fully independent goals, serializable goals (where there is an ordering for solving the goals without violating the previously solved ones), and arbitrarily interacting goals.”
Natural Language Processing With Modular PDP Networks and Distributed Lexicon (1991, background): “In general, building the system from hierarchically organized separable modules provides two major advantages: (1) the task is e ectively divided into subgoals, which is an e cient way to reduce complexity (Korf, 1987; Minsky, 1963), and (2) the system is forced to develop meaningful internal repres”
SHOP: Simple Hierarchical Ordered Planner (1999, background): “Problem reduction can be much more efficient than state-space search (even by an exponential amount in some cases [Korf, 1987; Yang et al 1992]).”
Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration (2020, background): “, 2012) and regression planning (RP) (Korf, 1987) to find a plan for a given goal based on its belief.”
The Fast Downward Planning System (2006, background): “In particular, search must fail in planning tasks which are not serializable (Korf, 1987).”