Hill climbing
Algorithms AIMech:HillClimbing DRAFT
Aliases: Hill-climbing search, Greedy local search
A family of local-improvement search procedures that replaces a current solution with a strictly better neighbor and stops when no improving neighbor exists under the specified move set.
Identity Notes & Concept Scope
The record covers the strict-improvement hill-climbing family: Glover selects some improving move, while the inspected AIMA figure selects the best-valued successor. Steepest ascent/descent is a particular selection rule, not an exact alias for every family member. Hill climbing does not denote every local-search metaheuristic, continuous gradient descent, greedy global best-first search, or the full ParamILS framework. No normalized identity match was found across all 442 pinned AIO labels/scoped synonyms and 304 preceding native IDs/names/aliases. Full-checkout fixed-string searches included ignored and hidden files in worktree and main, returning 176 and 176 lines respectively. Inspected text hits were prior discovery leads, uses of local search inside existing methods, source references and their saved reports/reviews. These were references to methods rather than existing candidate records. Six exact AIO CLASS nodes with scoped synonyms/comments and outgoing edges and six native neighboring mechanisms were inspected. Absence is bounded to this snapshot and semantic neighborhood.
Claims & Grounded Evidence
1 Claim
Mechanism
#method
Choose an initial solution, an objective and applicable neighborhood moves. In the minimizing formulation, select a move s with c(s(x))<c(x), replace x by s(x), and repeat. If no such move exists in the specified neighborhood, stop at the current solution. The source permits selecting any improving move. The inspected AIMA maximizing variant instead selects a highest-valued successor and stops if its value is no greater than the current value. Thus best-improvement and other improving-move choices share the strict-improvement principle but need not follow the same trajectory. A complete neighborhood check is needed to certify the stated local stopping condition; failure to find an improvement in a sample is not that certificate.
⚠️ Curated Limitations & Negative Results
- A local optimum need not be globally optimal. Strict improvement cannot traverse equal-valued plateaus or take an initially worsening escape step; sideways moves, random restarts and worsening acceptance change the specified baseline.
- Curator deduction: on a finite state space with a fixed objective and strict improvements, accepted states cannot repeat. This does not supply a polynomial bound, finite termination in arbitrary continuous spaces, or global optimality.
- Move definition, neighborhood coverage, initialization and selection/tie rules affect results. AIMA Figure 4.2 does not explicitly handle an empty successor set; a concrete implementation must declare that boundary. No numerical performance claim is curated.
Artificial Intelligence Ontology (AIO) Grounding
Explicit AIO Snapshot Assessment: No Match
Rationale: No suitable AIO target was found for this general neighborhood-search method within the inspected snapshot and semantic neighborhood. Optional model-training applications or objective evaluation do not establish a learning-task or model identity.
Search Scope: All 442 pinned AIO identities/scoped synonyms and 304 prior native IDs/names/aliases; ignored/hidden text in both checkouts; semantic label search for search, gradient and optimization; six exact AIO nodes and six native neighbors.
| Examined AIO Candidate | Rejection Reason |
|---|---|
AIO:TrainingStrategy
|
The definition concerns preprocessing or training machine-learning models. General local-search optimization is not restricted to such training. |
AIO:MachineLearningTask
|
The target concerns learning methods from data; minimizing a supplied combinatorial objective is not necessarily that task. |
AIO:ReinforcementLearning
|
Memory, adaptive penalties or an acceptance criterion do not themselves define a policy/environment learning problem. Exploration terminology is insufficient for a forced mapping. |
AIO:ReinforcementLearningLLM
|
The method is not a language model fine-tuned through reward feedback. |
AIO:EvolutionaryLLM
|
The target is a large language model. Its related evolutionary-algorithms synonym does not identify all optimizers or convert a single-trajectory method into a model. |
AIO:MarkovChain
|
Some memoryless configurations admit a Markov description, while others use search history. That mathematical representation alone does not identify the optimization framework with the pinned network/model class or its semantic parent. |
Evidence Sources & Primary Literature
2 Sources| Source ID | Publication / Document | Type / Access | Reference & Checksum |
|---|---|---|---|
method-source
|
Fred Glover
|
primary paper
full_text
|
ID: doi:10.1287/ijoc.1.3.190
sha256:fb1102afc415…
|
additional-1
|
Stuart Russell, Peter Norvig
|
official documentation
documentation
|
sha256:64a11241eb33…
|
Curation Provenance
Curator: Codex
Method: ai_assisted
Model: GPT-6
Created: 2026-10-11T14:27:03Z
Updated: 2026-10-11T14:30:11Z
Source File:
data/curated/HillClimbing.yaml