Beyond classical planning: temporal and probabilistic extensions
Bibliographic record
Abstract
Automated planning is a research field at the heart of artificial intelligence that deals with the central question of how to determine adequate actions that transform an initial state into a goal state. In classical planning, actions are deterministic and occur at single instant time points. Preconditions and effects are formulated over finite domain variables. To increase the applicability of planning for real world scenarios, we investigate two orthogonal extensions of classical planning. <br> <br>In the first part, we relax the restriction that actions have to occur at single instant time points in favor of a model that allows for durative actions that might overlap with each other. Furthermore, we consider numeric state variables that allow to represent real-valued fluents like fuel or money in a natural way. We present the temporal numeric planning system Temporal Fast Downward (TFD). Using a wide range of benchmarks we show that TFD is superior to state-of-the-art temporal planning systems in terms of both coverage and solution quality. <br> <br>The second part is concerned with probabilistic features. We drop the assumption of actions being deterministic and allow the definition of probability distributions over action outcomes. We show that it is worth to take probabilities into account in the decision making process by presenting several probability-aware algorithms for the Canadian Traveler's Problem, a challenging probabilistic planning benchmark. The demonstrated techniques clearly outperform the state-of-the-art approach that optimistically plans on a determinization of the original problem and replans in case of failure. We transfer the developed techniques to the more general field of domain-independent probabilistic planning. We present the probabilistic planning system PROST, winner of the most recent probabilistic planning competition. <br>
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.013 | 0.006 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.026 | 0.011 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".