scholarly journals Simple Deterministically Constructed Cycle Reservoirs with Regular Jumps

2012 ◽  
Vol 24 (7) ◽  
pp. 1822-1852 ◽  
Author(s):  
Ali Rodan ◽  
Peter Tiňo
Keyword(s):  
State Space ◽  
Unit Disk ◽  
Short Term Memory ◽  
Model Performance ◽  
Memory Capacity ◽  
Eigenvalue Distribution ◽  
Superior Performance ◽  
Reservoir Models ◽  
Linear Reservoir ◽  
High Degree

A new class of state-space models, reservoir models, with a fixed state transition structure (the “reservoir”) and an adaptable readout from the state space, has recently emerged as a way for time series processing and modeling. Echo state network (ESN) is one of the simplest, yet powerful, reservoir models. ESN models are generally constructed in a randomized manner. In our previous study (Rodan & Tiňo, 2011 ), we showed that a very simple, cyclic, deterministically generated reservoir can yield performance competitive with standard ESN. In this contribution, we extend our previous study in three aspects. First, we introduce a novel simple deterministic reservoir model, cycle reservoir with jumps (CRJ), with highly constrained weight values, that has superior performance to standard ESN on a variety of temporal tasks of different origin and characteristics. Second, we elaborate on the possible link between reservoir characterizations, such as eigenvalue distribution of the reservoir matrix or pseudo-Lyapunov exponent of the input-driven reservoir dynamics, and the model performance. It has been suggested that a uniform coverage of the unit disk by such eigenvalues can lead to superior model performance. We show that despite highly constrained eigenvalue distribution, CRJ consistently outperforms ESN (which has much more uniform eigenvalue coverage of the unit disk). Also, unlike in the case of ESN, pseudo-Lyapunov exponents of the selected optimal CRJ models are consistently negative. Third, we present a new framework for determining the short-term memory capacity of linear reservoir models to a high degree of precision. Using the framework, we study the effect of shortcut connections in the CRJ reservoir topology on its memory capacity.

scholarly journals Predicting Future Occurrence of Acute Hypotensive Episodes Using Noninvasive and Invasive Features

2021 ◽  
Vol 186 (Supplement_1) ◽  
pp. 445-451
Author(s):  
Yifei Sun ◽  
Navid Rashedi ◽  
Vikrant Vaze ◽  
Parikshit Shah ◽  
Ryan Halter ◽  
...  
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Real World ◽  
Short Term Memory ◽  
Model Performance ◽  
Learning Technologies ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Continuous Map

ABSTRACT Introduction Early prediction of the acute hypotensive episode (AHE) in critically ill patients has the potential to improve outcomes. In this study, we apply different machine learning algorithms to the MIMIC III Physionet dataset, containing more than 60,000 real-world intensive care unit records, to test commonly used machine learning technologies and compare their performances. Materials and Methods Five classification methods including K-nearest neighbor, logistic regression, support vector machine, random forest, and a deep learning method called long short-term memory are applied to predict an AHE 30 minutes in advance. An analysis comparing model performance when including versus excluding invasive features was conducted. To further study the pattern of the underlying mean arterial pressure (MAP), we apply a regression method to predict the continuous MAP values using linear regression over the next 60 minutes. Results Support vector machine yields the best performance in terms of recall (84%). Including the invasive features in the classification improves the performance significantly with both recall and precision increasing by more than 20 percentage points. We were able to predict the MAP with a root mean square error (a frequently used measure of the differences between the predicted values and the observed values) of 10 mmHg 60 minutes in the future. After converting continuous MAP predictions into AHE binary predictions, we achieve a 91% recall and 68% precision. In addition to predicting AHE, the MAP predictions provide clinically useful information regarding the timing and severity of the AHE occurrence. Conclusion We were able to predict AHE with precision and recall above 80% 30 minutes in advance with the large real-world dataset. The prediction of regression model can provide a more fine-grained, interpretable signal to practitioners. Model performance is improved by the inclusion of invasive features in predicting AHE, when compared to predicting the AHE based on only the available, restricted set of noninvasive technologies. This demonstrates the importance of exploring more noninvasive technologies for AHE prediction.

scholarly journals How Do Subway Signs Affect Pedestrians’ Wayfinding Behavior through Visual Short-Term Memory?

2021 ◽  
Vol 13 (12) ◽  
pp. 6866
Author(s):  
Haoru Li ◽  
Jinliang Xu ◽  
Xiaodong Zhang ◽  
Fangchen Ma
Keyword(s):  
Public Transportation ◽  
Short Term Memory ◽  
Memory Capacity ◽  
Cognitive Test ◽  
Subway Station ◽  
Short Term ◽  
Term Memory ◽  
Visual Short Term Memory ◽  
The Difference ◽  
Types Of Information

Recently, subways have become an important part of public transportation and have developed rapidly in China. In the subway station setting, pedestrians mainly rely on visual short-term memory to obtain information on how to travel. This research aimed to explore the short-term memory capacities and the difference in short-term memory for different information for Chinese passengers regarding subway signs. Previous research has shown that people’s general short-term memory capacity is approximately four objects and that, the more complex the information, the lower people’s memory capacity. However, research on the short-term memory characteristics of pedestrians for subway signs is scarce. Hence, based on the STM theory and using 32 subway signs as stimuli, we recruited 120 subjects to conduct a cognitive test. The results showed that passengers had a different memory accuracy for different types of information in the signs. They were more accurate regarding line number and arrow, followed by location/text information, logos, and orientation. Meanwhile, information type, quantity, and complexity had significant effects on pedestrians’ short-term memory capacity. Finally, according to our results that outline the characteristics of short-term memory for subway signs, we put forward some suggestions for subway signs. The findings will be effective in helping designers and managers improve the quality of subway station services as well as promoting the development of pedestrian traffic in such a setting.

Knowledge Enhanced LSTM for Coreference Resolution on Biomedical Texts

2021 ◽  
Author(s):  
Yufei Li ◽  
Xiaoyong Ma ◽  
Xiangyu Zhou ◽  
Pengzhen Cheng ◽  
Kai He ◽  
...  
Keyword(s):  
Information Integration ◽  
Short Term Memory ◽  
Superior Performance ◽  
Supplementary Information ◽  
Specific Information ◽  
Coreference Resolution ◽  
Fine Grained ◽  
Domain Specific ◽  
Memory Network ◽  
Biomedical Texts

Abstract Motivation Bio-entity Coreference Resolution focuses on identifying the coreferential links in biomedical texts, which is crucial to complete bio-events’ attributes and interconnect events into bio-networks. Previously, as one of the most powerful tools, deep neural network-based general domain systems are applied to the biomedical domain with domain-specific information integration. However, such methods may raise much noise due to its insufficiency of combining context and complex domain-specific information. Results In this paper, we explore how to leverage the external knowledge base in a fine-grained way to better resolve coreference by introducing a knowledge-enhanced Long Short Term Memory network (LSTM), which is more flexible to encode the knowledge information inside the LSTM. Moreover, we further propose a knowledge attention module to extract informative knowledge effectively based on contexts. The experimental results on the BioNLP and CRAFT datasets achieve state-of-the-art performance, with a gain of 7.5 F1 on BioNLP and 10.6 F1 on CRAFT. Additional experiments also demonstrate superior performance on the cross-sentence coreferences. Supplementary information Supplementary data are available at Bioinformatics online.

Mediation or Production Deficiency in Disabled Readers?

1980 ◽  
Vol 50 (2) ◽  
pp. 519-530
Author(s):  
Lauren Leslie
Keyword(s):  
Short Term Memory ◽  
Fifth Graders ◽  
Memory Capacity ◽  
Second Graders ◽  
Short Term ◽  
Term Memory ◽  
Memory Processing ◽  
Overt Rehearsal ◽  
Covert Rehearsal ◽  
Disabled Readers

Deficiencies in disabled readers’ short-term memory processing were studied. A deficit in memory capacity versus susceptibility to interference was investigated by examining performance over trials. A mediation versus production deficiency in memory processing was examined by testing the effect of instructions for rehearsal on performance of average and disabled readers in Grades 2 and 5. Contrary to prior research, facilitative effects of rehearsal instructions on second graders’ memory were found only on Trial 1. Fifth graders’ memory was adversely affected by overt rehearsal. Requiring children to rehearse overtly at a set rate may account for the results. A second study examined effects of covert rehearsal on the memory of average and disabled readers in Grade 2 over trials. Facilitative effects of covert rehearsal were shown when data of children who spontaneously rehearsed were removed. A deficiency in production by second graders was supported. Disabled readers who did not rehearse were more susceptible to interference.

scholarly journals Working memory as a factor mediating explicit and implicit knowledge of English grammar

2021 ◽  
pp. 1-8
Author(s):  
Mirosław Pawlak ◽  
Adriana Biedroń
Keyword(s):  
Working Memory ◽  
Short Term Memory ◽  
Working Memory Capacity ◽  
Memory Capacity ◽  
Target Language ◽  
Short Term ◽  
Passive Voice ◽  
English Grammar ◽  
Explicit And Implicit Knowledge ◽  
The Relationship

Abstract This paper reports the findings of a study that investigated the relationship between phonological short-term memory (PSTM), working memory capacity (WMC), and the level of mastery of L2 grammar. Grammatical mastery was operationalized as the ability to produce and comprehend English passive voice with reference to explicit and implicit (or highly automatized) knowledge. Correlational analysis showed that PSTM was related to implicit productive knowledge while WMC was linked to explicit productive knowledge. However, regression analysis showed that those relationships were weak and mediated by overall mastery of target language grammar, operationalized as final grades in a grammar course.

scholarly journals FetNet: a recurrent convolutional network for occlusion identification in fetoscopic videos

2020 ◽  
Vol 15 (5) ◽  
pp. 791-801 ◽  
Author(s):  
Sophia Bano ◽  
Francisco Vasconcelos ◽  
Emmanuel Vander Poorten ◽  
Tom Vercauteren ◽  
Sebastien Ourselin ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Laser Photocoagulation ◽  
Short Term Memory ◽  
Total Duration ◽  
Frame Rate ◽  
Superior Performance ◽  
Computer Assisted ◽  
Convolutional Network ◽  
Spatio Temporal

Abstract Purpose Fetoscopic laser photocoagulation is a minimally invasive surgery for the treatment of twin-to-twin transfusion syndrome (TTTS). By using a lens/fibre-optic scope, inserted into the amniotic cavity, the abnormal placental vascular anastomoses are identified and ablated to regulate blood flow to both fetuses. Limited field-of-view, occlusions due to fetus presence and low visibility make it difficult to identify all vascular anastomoses. Automatic computer-assisted techniques may provide better understanding of the anatomical structure during surgery for risk-free laser photocoagulation and may facilitate in improving mosaics from fetoscopic videos. Methods We propose FetNet, a combined convolutional neural network (CNN) and long short-term memory (LSTM) recurrent neural network architecture for the spatio-temporal identification of fetoscopic events. We adapt an existing CNN architecture for spatial feature extraction and integrated it with the LSTM network for end-to-end spatio-temporal inference. We introduce differential learning rates during the model training to effectively utilising the pre-trained CNN weights. This may support computer-assisted interventions (CAI) during fetoscopic laser photocoagulation. Results We perform quantitative evaluation of our method using 7 in vivo fetoscopic videos captured from different human TTTS cases. The total duration of these videos was 5551 s (138,780 frames). To test the robustness of the proposed approach, we perform 7-fold cross-validation where each video is treated as a hold-out or test set and training is performed using the remaining videos. Conclusion FetNet achieved superior performance compared to the existing CNN-based methods and provided improved inference because of the spatio-temporal information modelling. Online testing of FetNet, using a Tesla V100-DGXS-32GB GPU, achieved a frame rate of 114 fps. These results show that our method could potentially provide a real-time solution for CAI and automating occlusion and photocoagulation identification during fetoscopic procedures.

scholarly journals How high is visual short-term memory capacity for object layout?

2010 ◽  
Vol 72 (4) ◽  
pp. 1097-1109 ◽  
Author(s):  
Thomas Sanocki ◽  
Eric Sellers ◽  
Jeff Mittelstadt ◽  
Noah Sulman
Keyword(s):  
Short Term Memory ◽  
Memory Capacity ◽  
Short Term ◽  
Term Memory ◽  
Visual Short Term Memory

Learning written word vocabulary in a second language: Theoretical and practical implications

2017 ◽  
Vol 21 (3) ◽  
pp. 585-597 ◽  
Author(s):  
BRENDAN STUART WEEKES
Keyword(s):  
Second Language ◽  
Lexical Decision ◽  
Language Learning ◽  
Executive Control ◽  
Short Term Memory ◽  
Serial Order ◽  
Memory Capacity ◽  
Written Word ◽  
One Year ◽  
Serial Order Memory

Short-term memory (STM) is required for second language learning. However, it is not clear what components of STM are necessary for the acquisition and lexicalisation of new written words. Studies suggest that memory for serial order is a critical cognitive process in spoken word acquisition although correlated mechanisms such as executive control also play a role. In this study, bilingual Cantonese–English speakers who are learning written expert words in a non-native language were tested over a one year period in their first year of instruction. Written word lexicalisation was measured using lexical decision and spelling to dictation tasks. Results showed measures of executive control (Stroop performance) and serial order memory capacity predict recognition and recall of written expert words at different stages. Whereas serial order memory predicts improvements to lexical decision accuracy, executive control predicts spelling to dictation performance after one year. The conclusion is that STM processes do constrain written word lexicalisation in a second language. However, executive control and serial order memory capacity have differential effects during word lexicalisation.

Sign in / Sign up

Export Citation Format

Share Document