Bibliographic record
Abstract
Grammatical inference is a branch of computational learning theory that attacks the problem of learning grammatical models from string samples.In other words, grammatical inference tries to identify the computational models that generate the sample strings.In recent decade, due to the explosion of information, efficient ways of searching and organizing are of great importance.Therefore, using grammatical inference methods to process information computationally is very interesting.In real applications, the space-cost plays an important role on performance.In this thesis, we address the problem of refining learning models in grammatical inference.For regular language learning, we introduce formally the notion of "Classification Automaton" that reduces model size by identifying one automaton for multiple string classes.Classification automaton is proved to reduce 30% model size from a straightforward multiple automata approach on house rent data obtained from the public folder in Microsoft Exchange Server of Nanyang Technological University.In real world applications, there is always a maximum possible length for the strings.Based on this observation, we further introduce cover automata, which simplified a learning model with a maximum length limit, for grammatical inference.Test results based on Splice-junction Gene Sequence database demonstrate the method reduces model size by 32% from the widely used deterministic finite automaton model.By mixed k-th order Markov Chains, stochastic information for all possible substrings within k+1 length is captured.However, the space cost is exponential.We introduce the use of recurrent neural networks (RNNs) and present a pruning learning method to avoid the exponential space costs.There is a tradeoff between the accuracy and
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.023 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.006 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".