deepxube and DeepCubeAI
These are complementary approaches to learning neural network heuristics for combinatorial search problems, with DeepCubeAI using discrete world models while DeepXube learns domain-specific heuristics directly, allowing practitioners to combine both techniques or choose based on whether explicit state transition modeling is beneficial for their pathfinding domain.
About deepxube
forestagostinelli/deepxube
Learn a domain-specific heuristic function in a domain-independent fashion to solve pathfinding problems.
This project helps operations engineers and logistics planners find the most efficient paths for movement, deliveries, or resource allocation within complex systems. It takes a description of your system (the 'domain') and a starting point with a desired goal, then produces the optimal sequence of actions to reach that goal. This is designed for anyone needing to optimize navigation or sequencing tasks, like route planning for robots in a warehouse or determining steps in a manufacturing process.
About DeepCubeAI
misaghsoltani/DeepCubeAI
Learning Discrete World Models for Heuristic Search
This algorithm helps solve complex sequential decision-making problems, like those found in advanced puzzles or logistics. It takes an initial problem state and a desired goal state as input, then determines the optimal sequence of actions to reach that goal. This is ideal for researchers and practitioners in fields requiring intelligent agents to navigate and solve problems in discrete environments.
Scores updated daily from GitHub, PyPI, and npm data. How scores work