An active learning framework with a class \nbalancing strategy for time series classification
Bibliographic record
Abstract
Training machine learning models for classification tasks often requires labeling numerous \nsamples, which is costly and time-consuming, especially in time series analysis. \nThis research investigates Active Learning (AL) strategies to reduce the amount of \nlabeled data needed for e↵ective time series classification. Traditional AL techniques \ncannot control the selection of instances per class for labeling, leading to potential bias \nin classification performance and instance selection, particularly in imbalanced time \nseries datasets. To address this, we propose a novel class-balancing instance selection \nalgorithm integrated with standard AL strategies. Our approach aims to select more \ninstances from classes with fewer labeled examples, thereby addressing imbalance in \ntime series datasets. We demonstrate the e↵ectiveness of our AL framework in selecting \ninformative data samples for two distinct domains of tactile texture recognition \nand industrial fault detection. In robotics, our method achieves high-performance \ntexture categorization while significantly reducing labeled training data requirements \nto 70%. We also evaluate the impact of di↵erent sliding window time intervals on \nrobotic texture classification using AL strategies. In synthetic fiber manufacturing, \nwe adapt AL techniques to address the challenge of fault classification, aiming to \nminimize data annotation cost and time for industries. We also address real-life class \nimbalances in the multiclass industrial anomalous dataset using our class-balancing \ninstance algorithm integrated with AL strategies. Overall, this thesis highlights the \npotential of our AL framework across these two distinct domains.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.005 | 0.002 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 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".