Modifiability of Generalization in Dynamics Learning
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Studies on plasticity in motor function have shown that motor learning generalizes, such that movements in novel situations are affected by previous training. It has been shown that the pattern of generalization for visuomotor rotation learning changes when training movements are made to a wide distribution of directions. Here we have found that for dynamics learning, the shape of the generalization gradient is not similarly modifiable by the extent of training within the workspace. Subjects learned to control a robotic device during training and we measured how subsequent movements in a reference direction were affected. Our results show that as the angular separation between training and test directions increased, the extent of generalization was reduced. When training involved multiple targets throughout the workspace, the extent of generalization was no greater than following training to the nearest target alone. Thus a wide range of experience compensating for a dynamics perturbation provided no greater benefit than localized training. Instead, generalization was complete when training involved targets that bounded the reference direction. This suggests that broad generalization of dynamics learning to movements in novel directions depends on interpolation between instances of localized learning.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it