Bibliographic record
Abstract
While redundancy has been shown to improve accuracy and create a more robust network against system failures, it is still often an automatic assumption that redundancy is bad in neural networks. But are there more reasons to keep redundancy in neural networks? Can redundancy help neural networks perform different tasks and increase the efficiency of the network's performance? This paper investigates the role of redundancy in recurrent associative memory (RAM) by comparing redundant RAM networks to non-redundant RAM networks along the lines of learning, recall, and the ability to perform pattern segmentation. A bi-directional heteroassociative memory (BHM) recurrent neural network is used as the base network and modified to produce a network with redundancy. The amount of redundancy, memory load, noise in recall, and pattern segmentation ability are all set to varying levels to get the comprehensive picture of how redundancy affects the network. Ultimately, redundancy does have a purpose if there are weight differences for the input going into the redundant networks, i.e., there is redundancy in network structure, but these individual networks contain unique values for the input. The improvement is especially noticeable in the task of pattern segmentation. Having neural networks perform more complex tasks may provide even more incentive to incorporate redundancy into the network architecture.
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 | 0.007 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.005 |
| Scholarly communication | 0.003 | 0.010 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| 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".