Astus prunes the episodic-tree interpolation search using two pre-computed methods to keep error diagnosis tractable despite exponential growth of interpolated branches in the worst case¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q1
Currency · newest study 2015 · not yet checked against its sources · not reviewed by a person

Subclaims¶

Reading the evidence codes

q is the design tier, i the impact, n the sample, and the kind with its r the rigour judged within that kind: what each code means.

This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.

q1 i? In the worst case, the number of interpolated branches can increase exponentially for every interpolated procedure, motivating pruning of the interpolation tree. → Luc Paquette 2015 q1 i? Two pruning methods relying on information pre-computed from the procedural graph are implemented: accessible-primitive-procedure sets and data-flow analysis of queries. → Luc Paquette 2015 (2)

Evidence¶

Luc Paquette 2015¶

Luc Paquette, Jean-François Lebeau, Gabriel Beaulieu & André Mayers. (2015). Designing a Knowledge Representation Approach for the Generation of Pedagogical Interventions by MTTs. International Journal of Artificial Intelligence in Education 25:118–156. https://doi.org/10.1007/s40593-014-0030-z

q1 · i? · design · r2

Analytical complexity statement in the Episodic Tree Search Optimization section. Because interpolations are applied recursively to every procedure and to every combination of possible query disruptions, the article notes that "the number of interpolated branch can increase exponentially for every interpolated procedure," motivating pruning of branches that cannot lead to the learner's off-path step.

"In the worst case scenario, the number of interpolated branch can increase exponentially for every interpolated procedure."

Luc Paquette 2015 (2)¶

q1 · i? · design · r2

Description of the two implemented pruning methods in the Episodic Tree Search Optimization section. The first pre-computes, for each goal and procedure, "a set of primitive procedures that are accessible from that knowledge unit"; the second uses data-flow analysis to avoid disrupting queries unrelated to the off-path step's arguments.

"for every goal and procedure contained in the procedural graph, we compute a set of primitive procedures that are accessible from that knowledge unit."

Discussion¶

Astus diagnoses many learner errors without erroneous knowledge units by interpolating incorrect executions of procedures in the episodic tree — related

Page metadata

Field	Value
`type`	`claim`
`id`	`episodic-tree-search-pruning-optimizations`
`title`	`Astus prunes the episodic-tree interpolation search using two pre-computed methods to keep error diagnosis tractable despite exponential growth of interpolated branches in the worst case`
`description`	`Astus prunes the episodic-tree interpolation search using two pre-computed methods to keep error diagnosis tractable despite exponential growth of interpolated branches in the worst case`
`status`	`review`
`evidence_strength`	`weak`
`newest_evidence`	`2015`
`generated`	by `process:wiki-ingest` on `2026-10-27`
`trust`	unverified — no one has confirmed this page's content
`sources`	2 — `luc-paquette-2015`, `luc-paquette-2015-2`

Frontmatter is the page's machine-readable record — see the Schema & Guide.