Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

How do PDP models learn quasiregularity?

ABSTRACT: Parallel distributed processing (PDP) models have had a profound impact on the study of cognition. One domain in which they have been particularly influential is learning quasiregularity, in which mastery requires both learning regularities that capture the majority of the structure in the input plus learning exceptions that violate the regularities. How PDP models learn quasiregularity is still not well understood. Small- and large-scale analyses of a feedforward, 3-layer network were carried out to address 2 fundamental issues about network functioning: how the model can learn both regularities and exceptions without sacrificing generalizability and the nature of the hidden representation that makes this learning possible. Results show that capacity-limited learning pressures the network to form componential representations, which ensures good generalizability. Small and highly local perturbations of this representational system allow exceptions to be learned while minimally disrupting generalizability. Theoretical and methodological implications of the findings are discussed.

SUBMITTER: Kim W

PROVIDER: S-EPMC4029953 | biostudies-literature | 2013 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

How do PDP models learn quasiregularity?

Kim Woojae W Pitt Mark A MA Myung Jay I JI

Psychological review 20130909 4

Parallel distributed processing (PDP) models have had a profound impact on the study of cognition. One domain in which they have been particularly influential is learning quasiregularity, in which mastery requires both learning regularities that capture the majority of the structure in the input plus learning exceptions that violate the regularities. How PDP models learn quasiregularity is still not well understood. Small- and large-scale analyses of a feedforward, 3-layer network were carried o ...[more]

PMID: 24015955

Similar Datasets

Language models can learn complex molecular distributions.

Project description:Deep generative models of molecules have grown immensely in popularity, trained on relevant datasets, these models are used to search through chemical space. The downstream utility of generative models for the inverse design of novel functional compounds, depends on their ability to learn a training distribution of molecules. The most simple example is a language model that takes the form of a recurrent neural network and generates molecules using a string representation. Since their initial use, subsequent work has shown that language models are very capable, in particular, recent research has demonstrated their utility in the low data regime. In this work, we investigate the capacity of simple language models to learn more complex distributions of molecules. For this purpose, we introduce several challenging generative modeling tasks by compiling larger, more complex distributions of molecules and we evaluate the ability of language models on each task. The results demonstrate that language models are powerful generative models, capable of adeptly learning complex molecular distributions. Language models can accurately generate: distributions of the highest scoring penalized LogP molecules in ZINC15, multi-modal molecular distributions as well as the largest molecules in PubChem. The results highlight the limitations of some of the most popular and recent graph generative models- many of which cannot scale to these molecular distributions.

| S-EPMC9174447 | biostudies-literature

Models that learn how humans learn: The case of decision-making and its disorders.

Project description:Popular computational models of decision-making make specific assumptions about learning processes that may cause them to underfit observed behaviours. Here we suggest an alternative method using recurrent neural networks (RNNs) to generate a flexible family of models that have sufficient capacity to represent the complex learning and decision- making strategies used by humans. In this approach, an RNN is trained to predict the next action that a subject will take in a decision-making task and, in this way, learns to imitate the processes underlying subjects' choices and their learning abilities. We demonstrate the benefits of this approach using a new dataset drawn from patients with either unipolar (n = 34) or bipolar (n = 33) depression and matched healthy controls (n = 34) making decisions on a two-armed bandit task. The results indicate that this new approach is better than baseline reinforcement-learning methods in terms of overall performance and its capacity to predict subjects' choices. We show that the model can be interpreted using off-policy simulations and thereby provides a novel clustering of subjects' learning processes-something that often eludes traditional approaches to modelling and behavioural analysis.

| S-EPMC6588260 | biostudies-literature

Ag(2)PdP(2)O(7).

Project description:Disilver(I) palladium(II) diphosphate, Ag(2)PdP(2)O(7), is isotypic with Na(2)PdP(2)O(7). It consists of infinite diphosphato-pallad-ate(II) [Pd(P(2)O(7))(2/2)](2-) ribbons with the Pd(II) ion in an almost square-planar coordination ( symmetry) and the P(2)O(7) group exhibiting 2 symmetry. The [Pd(P(2)O(7))(2/2)](2-) ribbons are linked by distorted [AgO(6)] octa-hedra. (31)P-MAS NMR studies on Ag(2)PdP(2)O(7) are in accordance with one independent site for phospho-rus; its isotropic chemical shift δ(iso) = 21.5 p.p.m. is similar to that of Pd(2)P(2)O(7).

| S-EPMC2959806 | biostudies-literature

Improving the management procedures in farms infected with the Porcine Reproductive and Respiratory Syndrome virus using PDP models.

Project description:Pig meat production need to be built up in the future due to the increase of the human population worldwide. To address this challenge, there is plenty of room for improvement in terms of pig production efficiency that could be severely hampered by the presence of diseases. In this sense, Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) is one of the most costly disease present in industrial pork production in Europe and North America. We have developed a model to analyze the effect of different management procedures to control this important virus in different epidemiological scenarios. Our results clearly suggest that no cross-fostering during lactation and the maintaining of litter integrity significantly decrease the number of sick and dead animals during the rearing period compared to scenarios where cross-fostering and no litter integrity are practiced. These results highlight the relevance of different management strategies to control PRRSV and quantify the effect of limiting cross-fostering and avoiding mixing animals from different litters in PRRSV positive farms to optimize animal production. Our findings will allow pig farmers to apply these management procedures to control this disease under field conditions in a very cost-effective way.

| S-EPMC6620323 | biostudies-literature

Protein language models trained on multiple sequence alignments learn phylogenetic relationships.

Project description:Self-supervised neural language models with attention have recently been applied to biological sequence data, advancing structure, function and mutational effect prediction. Some protein language models, including MSA Transformer and AlphaFold's EvoFormer, take multiple sequence alignments (MSAs) of evolutionarily related proteins as inputs. Simple combinations of MSA Transformer's row attentions have led to state-of-the-art unsupervised structural contact prediction. We demonstrate that similarly simple, and universal, combinations of MSA Transformer's column attentions strongly correlate with Hamming distances between sequences in MSAs. Therefore, MSA-based language models encode detailed phylogenetic relationships. We further show that these models can separate coevolutionary signals encoding functional and structural constraints from phylogenetic correlations reflecting historical contingency. To assess this, we generate synthetic MSAs, either without or with phylogeny, from Potts models trained on natural MSAs. We find that unsupervised contact prediction is substantially more resilient to phylogenetic noise when using MSA Transformer versus inferred Potts models.

| S-EPMC9588007 | biostudies-literature

Candidatus Hepatincola sp. Pdp

Project description:Candidatus Hepatincola sp. Pdp Genome sequencing and assembly

| PRJNA690772 | ENA

In Vivo Models for Cholangiocarcinoma-What Can We Learn for Human Disease?

Project description:Cholangiocarcinoma (CCA) comprises a heterogeneous group of primary liver tumors. They emerge from different hepatic (progenitor) cell populations, typically via sporadic mutations. Chronic biliary inflammation, as seen in primary sclerosing cholangitis (PSC), may trigger CCA development. Although several efforts were made in the last decade to better understand the complex processes of biliary carcinogenesis, it was only recently that new therapeutic advances have been achieved. Animal models are a crucial bridge between in vitro findings on molecular or genetic alterations, pathophysiological understanding, and new therapeutic strategies for the clinic. Nevertheless, it is inherently difficult to recapitulate simultaneously the stromal microenvironment (e.g., immune-competent cells, cholestasis, inflammation, PSC-like changes, fibrosis) and the tumor biology (e.g., mutational burden, local growth, and metastatic spread) in an animal model, so that it would reflect the full clinical reality of CCA. In this review, we highlight available data on animal models for CCA. We discuss if and how these models reflect human disease and whether they can serve as a tool for understanding the pathogenesis, or for predicting a treatment response in patients. In addition, open issues for future developments will be discussed.

| S-EPMC7404171 | biostudies-literature

Distillation of crop models to learn plant physiology theories using machine learning.

Project description:Convolutional neural networks (CNNs) can not only classify images but can also generate key features, e.g., the Google neural network that learned to identify cats by simply watching YouTube videos, for the classification. In this paper, crop models are distilled by CNN to evaluate the ability of deep learning to identify the plant physiology knowledge behind such crop models simply by learning. Due to difficulty in collecting big data on crop growth, a crop model was used to generate datasets. The generated datasets were fed into CNN for distillation of the crop model. The models trained by CNN were evaluated by the visualization of saliency maps. In this study, three saliency maps were calculated using all datasets (case 1) and using datasets with spikelet sterility due to either high temperature at anthesis (case 2) or cool summer damage (case 3). The results of case 1 indicated that CNN determined the developmental index of paddy rice, which was implemented in the crop model, simply by learning. Moreover, CNN identified the important individual environmental factors affecting the grain yield. Although CNN had no prior knowledge of spikelet sterility, cases 2 and 3 indicated that CNN realized about paddy rice becoming sensitive to daily mean and maximum temperatures during specific periods. Such deep learning approaches can be used to accelerate the understanding of crop models and make the models more portable. Moreover, the results indicated that CNN can be used to develop new plant physiology theories simply by learning.

| S-EPMC6541271 | biostudies-literature

Comparing International Models of Integrated Care: How Can We Learn Across Borders?

Project description:IntroductionProviders, managers, health system leaders, and researchers could learn across countries implementing system-wide models of integrated care, but require accessible methods to do so. This study assesses if a common framework could describe and compare key components of international models of integrated care.Theory and methodsA framework developed for an international study of programs that address high needs high cost patients was used to describe and compare 11 case studies analyzed in two international research projects; the Implementing Integrated Care for Older Adults with Complex Health Needs (iCOACH) study in Canada and New Zealand, and the Vilans research group exploring models in the Netherlands. Comparative summaries were generated, with findings discussed at a 2019 International Conference on Integrated Care workshop.ResultsThe template was found to be useful to compare integrated case analyses in different contexts, and stands apart from other case comparison approaches as it is easily applied and can provide practical guidance for frontline staff and managers. Areas of improvement for the template are identified and two updated versions are presented.Conclusions and discussionThere is value to using a common template to provide guidance in international comparison of models of integrated care. We discuss the applicability of the approach to support scale and spread of integrated care internationally.

| S-EPMC7147684 | biostudies-literature