Fear generalization refers to the process, whereby a stimulus acquires aversive properties due to both its' perceptual and/or conceptual similarity to another aversive stimulus. Since excessive fear generalization can account for the burden in life experienced by many anxiety patients, research on fear...
Domain generalization refers to the ability of a model to perform well on data from unseen domains that differ from the training data distribution. Marginal domain generalization specifically focuses on handling marginal shifts in the data distribution, where small changes occur in the input features ...
Specifically, participants indicated their expectations regarding the variability of stimuli in the application set when temporal distance between learning and application was short versus long. Experiment 3 then replicated Experiment 1, but instead of direct instructions to anticipate more variability, it ...
Specifically, GB refers to a limitation that the generalization capacity of lower-dimensional embedding space is inferior to the higher-dimensional feature space in the test stage. To mitigate the capacity gap between feature space and embedding space, we propose to introduce a fully-learnable module...
(stimulus A) and “negative” (stimulus C) are paired. Then, if new bidirectional relations occur, “social” (stimulus B) and “negative” (stimulus C) might be related, even though that equivalence was never specifically taught. Such conditioning could then lead to generalization across social...
kicking off those with obvious extrapolation according to the mean extrapolation distance (MED). The second one is ranking the feature and determining the feature subset according to the model performance. The whole process is shown in Fig.4and to be more specifically, it includes the following ...
Use the format "dataset-numbers" to select fault categories, where 'numbers' refers to specific faults.For example, consider the sample datasets in the 'Releases' named Dataset-TL-FD-Library: Transfer from CWRU (inner, normal, and outer faults) to MFPT (inner and normal faults) dataset....
To increase the number of domains and support the flexible DG scenarios where the training do- mains are not aligned with respect to categories, we further attain unique domains specifically for each of the 80 cate- gories. We select the unique domains according to the...
Typically, a “domain” refers to the specific data type or category on which a model is trained. Here, we apply this concept to the annotation of cellular malignant states in single-cell or spatial data by assuming that data from different tissues arise from different domains. The generalized...
A second category of generalization studies focuses on structural generalization—the extent to which models can process or generate structurally (grammatically) correct output—rather than on whether they can assign them correct interpretations. Some structural generalization studies focus specifically on synt...