scholarly journals Memory-Based Semantic Parsing

2021 ◽  
Vol 9 ◽  
pp. 1197-1212
Author(s):  
Parag Jain ◽  
Mirella Lapata
Keyword(s):  
Contextual Information ◽  
Experimental Results ◽  
External Memory ◽  
Semantic Parsing ◽  
Memory Controller ◽  
Context Dependent ◽  
Improved Performance ◽  
Process Context ◽  
Context Memory

Abstract We present a memory-based model for context- dependent semantic parsing. Previous approaches focus on enabling the decoder to copy or modify the parse from the previous utterance, assuming there is a dependency between the current and previous parses. In this work, we propose to represent contextual information using an external memory. We learn a context memory controller that manages the memory by maintaining the cumulative meaning of sequential user utterances. We evaluate our approach on three semantic parsing benchmarks. Experimental results show that our model can better process context-dependent information and demonstrates improved performance without using task-specific decoders.

scholarly journals Patients with Schizophrenia Remember that an Event has Occurred, but not When

1996 ◽  
Vol 168 (4) ◽  
pp. 427-431 ◽  
Author(s):  
Lydia Rizzo ◽  
Jean-Marie Danion ◽  
Martial Van Der Linden ◽  
Danielle Grangé
Keyword(s):  
Memory Impairment ◽  
Discrimination Task ◽  
Contextual Information ◽  
Normal Subjects ◽  
Memory Deficit ◽  
Temporal Context ◽  
Context Memory

BackgroundThe context memory deficit hypothesis of schizophrenia postulates that the long-term deficit associated with this disorder is related to a memory impairment for contextual information.MethodTo test this hypothesis, memory for temporal context was assessed in 33 patients with schizophrenia and 33 normal subjects, using a recency discrimination task.ResultsWhereas patients were able to recall and recognise target items, they were unable to recognise from among the target items those which had been most recently learned.ConclusionsSchizophrenia is associated with a temporal context memory deficit.

Micro CO2 Sensor Based on Li2CO3, Pt | Li3PO4 | Pt, Li2TiO3-TiO2 Structure

2010 ◽  
Vol 44-47 ◽  
pp. 2747-2751
Author(s):  
Hai Rong Wang ◽  
Jun Qiang Ren ◽  
Guo Liang Sun ◽  
Di Cen
Keyword(s):  
Solid Electrolyte ◽  
Gas Sensor ◽  
Experimental Results ◽  
Al2o3 Substrate ◽  
Reference Electrodes ◽  
Co2 Gas ◽  
Speed Of Response ◽  
Co2 Gas Sensor ◽  
Improved Performance ◽  
Mems Process

This paper presents a micro solid electrolyte CO2 gas sensor in which Li2CO3, Li2TiO3-TiO2 serves as sensing and reference electrodes respectively, and the Li3PO4 film acts as the electrolyte. The sensor was constructed in the sequent layers of O2, CO2, Li2CO3, Pt | Li3PO4 | Pt, Li2TiO3-TiO2, O2, CO2 on the Al2O3 substrate by MEMS process. Experimental results indicate that the micro solid-electrolyte CO2 gas sensor has a relatively rapid speed of response. By discussions, we may find that the improved performance will be realized by optimizing the primary parameters of the sensor.

scholarly journals Fast Parallel All-Subgraph Enumeration Using Multicore Machines

2015 ◽  
Vol 2015 ◽  
pp. 1-11 ◽  
Author(s):  
Saeed Shahrivari ◽  
Saeed Jalili
Keyword(s):  
Complex Networks ◽  
Real World ◽  
Main Memory ◽  
Important Task ◽  
Experimental Results ◽  
External Memory ◽  
Input Graph ◽  
Enumeration Algorithm ◽  
Parallel Solution ◽  
Processor Cores

Enumerating all subgraphs of an input graph is an important task for analyzing complex networks. Valuable information can be extracted about the characteristics of the input graph using all-subgraph enumeration. Notwithstanding, the number of subgraphs grows exponentially with growth of the input graph or by increasing the size of the subgraphs to be enumerated. Hence, all-subgraph enumeration is very time consuming when the size of the subgraphs or the input graph is big. We propose a parallel solution namedSubenumwhich in contrast to available solutions can perform much faster. Subenum enumerates subgraphs using edges instead of vertices, and this approach leads to a parallel and load-balanced enumeration algorithm that can have efficient execution on current multicore and multiprocessor machines. Also, Subenum uses a fast heuristic which can effectively accelerate non-isomorphism subgraph enumeration. Subenum can efficiently use external memory, and unlike other subgraph enumeration methods, it is not associated with the main memory limits of the used machine. Hence, Subenum can handle large input graphs and subgraph sizes that other solutions cannot handle. Several experiments are done using real-world input graphs. Compared to the available solutions, Subenum can enumerate subgraphs several orders of magnitude faster and the experimental results show that the performance of Subenum scales almost linearly by using additional processor cores.

scholarly journals Encoding context determines risky choice

2017 ◽  
Author(s):  
Christopher R Madan ◽  
Marcia L Spetch ◽  
Fernanda Machado ◽  
Alice Mason ◽  
Elliot Andrew Ludvig
Keyword(s):  
Risky Choice ◽  
Local Context ◽  
Single Session ◽  
Memory Tests ◽  
Global Context ◽  
Risky Choices ◽  
Context Dependent ◽  
Context Memory ◽  
Encoding And Retrieval

Both memory and choice are influenced by context: Memory is enhanced when encoding and retrieval contexts match, and choice is swayed by available options. Here, we assessed how context influences risky choice in an experience-based task. Within a single session, we created two separate contexts by presenting blocks of trials in distinct backgrounds. Risky choices were context-dependent; given the same choice, people chose differently depending on other outcomes experienced in that context. Choices reflected an overweighting of the most extreme outcomes within each local context, rather than the global context of all outcomes. When tested in the non-trained context, people chose according to the context at encoding and not retrieval. In subsequent memory tests, people displayed biases specific to distinct contexts: extreme outcomes from each context were more accessible and judged as more frequent. These results pose a challenge for theories of choice that rely on retrieval as guiding choice.

scholarly journals The Left Angular Gyrus Is Causally Involved in Context-dependent Integration and Associative Encoding during Narrative Reading

2021 ◽  
pp. 1-14
Author(s):  
Francesca M. Branzi ◽  
Gorana Pobric ◽  
JeYoung Jung ◽  
Matthew A. Lambon Ralph
Keyword(s):  
Language Processing ◽  
Information Integration ◽  
Contextual Information ◽  
Memory Task ◽  
Causal Link ◽  
Angular Gyrus ◽  
Line Context ◽  
On Line ◽  
Context Dependent ◽  
Left Angular Gyrus

Abstract The role of the left angular gyrus (AG) in language processing remains unclear. In this study, we used TMS to test the hypothesis that the left AG causally supports the processes necessary for context-dependent integration and encoding of information during language processing. We applied on-line TMS over the left AG to disrupt the on-line context-dependent integration during a language reading task, specifically while human participants integrated information between two sequentially presented paragraphs of text (“context” and “target” paragraphs). We assessed the effect of TMS on the left AG by asking participants to retrieve integrated contextual information when given the target condition as cue in a successive memory task. Results from the memory task showed that TMS applied over the left AG during reading impaired the formation of integrated context-target representation. These results provide the first evidence of a causal link between the left AG function, on-line information integration, and associative encoding during language processing.

scholarly journals Aging, Culture, and Memory for Socially Meaningful Item-Context Associations: An East-West Cross-Cultural Comparison Study

2021 ◽  
Author(s):  
Lixia Yang ◽  
Juan Li ◽  
Julia Spaniol ◽  
Lynn Hasher ◽  
Andrea J. Wilkinson ◽  
...  
Keyword(s):  
Independent Living ◽  
Daily Life ◽  
Contextual Information ◽  
Cross Cultural ◽  
Cultural Comparison ◽  
Comparison Study ◽  
Cross Cultural Comparison ◽  
Age Related ◽  
East West ◽  
Context Memory

Research suggests that people in Eastern interdependent cultures process information more holistically and attend more to contextual information than do people in Western independent cultures. The current study examined the effects of culture and age on memory for socially meaningful item-context associations in 71 Canadians of Western European descent (35 young and 36 older) and 72 native Chinese citizens (36 young and 36 older). All participants completed two blocks of context memory tasks. During encoding, participants rated pictures of familiar objects. In one block, objects were rated either for their meaningfulness in the independent living context or their typicality in daily life. In the other block, objects were rated for their meaningfulness in the context of fostering relationships with others or for their typicality in daily life. The encoding in each block was followed by a recognition test in which participants identified pictures and their associated contexts. The results showed that Chinese outperformed Canadians in context memory, though both culture groups showed similar age-related deficits in item and context memory. The results suggest that Chinese are at an advantage in memory for socially meaningful item-context associations, an advantage that continues from young adulthood into old age.

scholarly journals Context-dependent decision making in a premotor circuit

2019 ◽  
Author(s):  
Zheng Wu ◽  
Ashok Litwin-Kumar ◽  
Philip Shamash ◽  
Alexei Taylor ◽  
Richard Axel ◽  
...  
Keyword(s):  
Pyramidal Neurons ◽  
Contextual Information ◽  
Sensory Information ◽  
Behavioral Responses ◽  
Association Cortex ◽  
Match To Sample ◽  
Subsequent Test ◽  
Layer 2 ◽  
Test Odor ◽  
Context Dependent

SummaryCognitive capacities afford contingent associations between sensory information and behavioral responses. We studied this problem using an olfactory delayed match to sample task whereby a sample odor specifies the association between a subsequent test odor and rewarding action. Multi-neuron recordings revealed representations of the sample and test odors in olfactory sensory and association cortex, which were sufficient to identify the test odor as match/non-match. Yet, inactivation of a downstream premotor area (ALM), but not orbitofrontal cortex, confined to the epoch preceding the test odor, led to gross impairment. Olfactory decisions that were not context dependent were unimpaired. Therefore, ALM may not receive the outcome of a match/non-match decision from upstream areas but contextual information—the identity of the sample—to establish the mapping between test odor and action. A novel population of pyramidal neurons in ALM layer 2 may mediate this process.

scholarly journals Weakly Supervised Learning for Object Localization Based on an Attention Mechanism

2021 ◽  
Vol 11 (22) ◽  
pp. 10953
Author(s):  
Nojin Park ◽  
Hanseok Ko
Keyword(s):  
Deep Learning ◽  
Experimental Results ◽  
Object Localization ◽  
Localization Accuracy ◽  
Large Size ◽  
Bounding Box ◽  
Improved Performance ◽  
Weakly Supervised ◽  
Detection And Localization ◽  
Activation Map

Recently, deep learning has been successfully applied to object detection and localization tasks in images. When setting up deep learning frameworks for supervised training with large datasets, strongly labeling the objects facilitates good performance; however, the complexity of the image scene and large size of the dataset make this a laborious task. Hence, it is of paramount importance that the expensive work associated with the tasks involving strong labeling, such as bounding box annotation, is reduced. In this paper, we propose a method to perform object localization tasks without bounding box annotation in the training process by means of employing a two-path activation-map-based classifier framework. In particular, we develop an activation-map-based framework to judicially control the attention map in the perception branch by adding a two-feature extractor so that better attention weights can be distributed to induce improved performance. The experimental results indicate that our method surpasses the performance of the existing deep learning models based on weakly supervised object localization. The experimental results show that the proposed method achieves the best performance, with 75.21% Top-1 classification accuracy and 55.15% Top-1 localization accuracy on the CUB-200-2011 dataset.

scholarly journals Korean comparative constructions: A constraint-based approach and computational implementation

2010 ◽  
Author(s):  
Jong-Bok Kim ◽  
Jaehyung Yang ◽  
Sanghoun Song
Keyword(s):  
Contextual Information ◽  
Semantic Representations ◽  
Feature Structure ◽  
Computational Implementation ◽  
Comparative Constructions ◽  
Context Dependent ◽  
Computational Analyses ◽  
Further Development

The complexity of comparative constructions in each language has given challenges to both theoretical and computational analyses. This paper first identifies types of comparative constructions in Korean and discusses their main grammatical properties. It then builds a syntactic parser couched upon the typed feature structure grammar, HPSG and proposes a context-dependent interpretation for the comparison. To check the feasibility of the proposed analysis, we have implemented the grammar into the existing Korean Resource Grammar. The results show us that the grammar we have developed here is feasible enough to parse Korean comparative sentences and yield proper semantic representations though further development is needed for a finer model for contextual information.

Optimization of Non-axisymmetric Endwall Contours for the Rotor of a Low Speed, 112-Stage Research Turbine With Unshrouded Blades—Optimization and Experimental Validation

2020 ◽  
Vol 142 (4) ◽  
Author(s):  
Jonathan Bergh ◽  
Glen Snedden ◽  
Dwain Dunn
Keyword(s):  
Experimental Results ◽  
Pressure Probe ◽  
Test Case ◽  
Current Case ◽  
Low Speed ◽  
Additional Information ◽  
Measurement Plane ◽  
Direct Laser ◽  
Improved Performance ◽  
Design And Practice

Abstract This paper presents the predicted, as well as final experimental results for the design of an automatically optimized non-axisymmetric endwall and as such, attempts to close the loop between design and practice, providing additional information to other groups involved in the design of endwall contours. The contours designed in this investigation were manufactured using the direct laser sintering rapid prototyping method and installed and tested in the low-speed, 112-stage turbine at the CSIR’s test turbine facility (TTF) in Pretoria, South Africa. Steady-state 5-hole pressure probe traverses were used to characterize the performance and flow profiles upstream, immediately downstream and in a quasi-“mixed-out” sense downstream of the rotor. In addition to the datum (annular) case, both the computed as well as experimental results were compared to the corresponding results generated for a “generically” contoured rotor which was originally designed for a linear cascade test case, but one which used the same blade profile to the current case. The results show that in general both sets of contours performed well, although the added emphasis on flow correction for the contours produced in this investigation resulted in slightly worse performance in terms of loss at the rotor exit (X3) but greatly improved performance in terms of the efficiency and flow angles at the “mixed-out” (X4) measurement plane.

Sign in / Sign up

Export Citation Format

Share Document