scholarly journals Continual Learning for Grounded Instruction Generation by Observing Human Following Behavior

2021 ◽  
Vol 9 ◽  
pp. 1303-1319
Author(s):  
Noriyuki Kojima ◽  
Alane Suhr ◽  
Yoav Artzi
Keyword(s):  
Natural Language ◽  
Original System ◽  
Language Instruction ◽  
Over Time ◽  
Continual Learning

Abstract We study continual learning for natural language instruction generation, by observing human users’ instruction execution. We focus on a collaborative scenario, where the system both acts and delegates tasks to human users using natural language. We compare user execution of generated instructions to the original system intent as an indication to the system’s success communicating its intent. We show how to use this signal to improve the system’s ability to generate instructions via contextual bandit learning. In interaction with real users, our system demonstrates dramatic improvements in its ability to generate language over time.

scholarly journals Joint Multimodal Embedding and Backtracking Search in Vision-and-Language Navigation

Sensors ◽  
2021 ◽  
Vol 21 (3) ◽  
pp. 1012
Author(s):  
Jisu Hwang ◽  
Incheol Kim
Keyword(s):  
Natural Language ◽  
Language Processing ◽  
Input Data ◽  
Language Instruction ◽  
Scoring Method ◽  
Processing Technologies ◽  
Backtracking Search ◽  
Panoramic Images ◽  
Benchmark Datasets ◽  
Vision And Language

Due to the development of computer vision and natural language processing technologies in recent years, there has been a growing interest in multimodal intelligent tasks that require the ability to concurrently understand various forms of input data such as images and text. Vision-and-language navigation (VLN) require the alignment and grounding of multimodal input data to enable real-time perception of the task status on panoramic images and natural language instruction. This study proposes a novel deep neural network model (JMEBS), with joint multimodal embedding and backtracking search for VLN tasks. The proposed JMEBS model uses a transformer-based joint multimodal embedding module. JMEBS uses both multimodal context and temporal context. It also employs backtracking-enabled greedy local search (BGLS), a novel algorithm with a backtracking feature designed to improve the task success rate and optimize the navigation path, based on the local and global scores related to candidate actions. A novel global scoring method is also used for performance improvement by comparing the partial trajectories searched thus far with a plurality of natural language instructions. The performance of the proposed model on various operations was then experimentally demonstrated and compared with other models using the Matterport3D Simulator and room-to-room (R2R) benchmark datasets.

Training personal robots using natural language instruction

2001 ◽  
Vol 16 (5) ◽  
pp. 38-45 ◽  
Author(s):  
S. Lauria ◽  
G. Bugmann ◽  
T. Kyriacou ◽  
J. Bos ◽  
A. Klein
Keyword(s):  
Natural Language ◽  
Language Instruction ◽  
Personal Robots

Evidence Based on a dynamic source: Database support for a theory of transitive reciprocals

2015 ◽  
Vol 1 (1) ◽  
Author(s):  
Ken Safir
Keyword(s):  
Natural Language ◽  
Empirical Support ◽  
Direct Object ◽  
Evidence Based ◽  
Change Over Time ◽  
Search Results ◽  
Source Database ◽  
Linguistic Research ◽  
Dynamic Source ◽  
Over Time

AbstractThe novelty of this document is that the empirical support for the predictions it examines, predictions about the distribution and interpretation of transitive reciprocal constructions, will be different each time it is read. The evidence will change because this paper will only provide parameters for a search of the Afranaph Database (ongoing) and two other databases, and as these databases grow and change over time, the search results returned today will be different from the results returned by the same search executed months or years from now. Reversing the normal priorities of linguistic research, the proposal we present about the nature of reciprocal constructions in natural language, which contends that direct object full DPs anaphors do not directly contribute reciprocal meaning (a proposal more broadly and specifically defended by

scholarly journals PERKEMBANGAN MAKNA PEYORATIF PADA KATA BAHASA INGGRIS ‘IDIOT’: SEBUAH KAJIAN ETIMOLOGI

2018 ◽  
Vol 7 (1) ◽  
pp. 31-38
Author(s):  
Nur Ajurah
Keyword(s):  
Natural Language ◽  
Mentally Ill ◽  
Qualitative Method ◽  
English Word ◽  
Semantic Features ◽  
Private Person ◽  
Original Meaning ◽  
History Of ◽  
Negative Connotation ◽  
Over Time

This study argues that every word of any natural language is prone to meaning modification known as pejoration. Pejoration happens when a meaning of words becomes negative and it is different from its original meaning. In order to answer that phenomenon, this study entitled “Pejorative Development of English Word ‘Idiot’: A Study of Etymology” issues that the English word ‘idiot’ may have experienced pejoration. It discusses the history of the word idiot’ and its pejorative development. Liberman and Voyles’s theories are used in this study. In the analysis, Liberman’s theory is applied to explain the history of the word ‘idiot’ and also the development of its meaning, while Voyles’s theory is adapted to see the semantic features of the word ‘idiot’ whether the meaning is specific or general.      The descriptive qualitative method is used to explain the phenomena covered in pejorative datum. The datum was analyzed and described using applying grand and supporting theories mentioned earlier.             The result shows that the word ‘idiot’ has  experienced degradation of meaning where the word itself in its original meaning meant ‘a private person’, but over time it has acquired a negative connotation. The word is currently used in the sense of ‘a stupid person or someone has done something stupid’ or ‘someone who is mentally ill or has a very low level of intelligence’.

scholarly journals Grounding of Human Environments and Activities for Autonomous Robots

2017 ◽  
Author(s):  
Muhannad Alomari ◽  
Paul Duckworth ◽  
Nils Bore ◽  
Majd Hawasly ◽  
David C. Hogg ◽  
...  
Keyword(s):  
Natural Language ◽  
Real World ◽  
Human Activities ◽  
Autonomous Robots ◽  
Proof Of Concept ◽  
Learning Framework ◽  
Art Object ◽  
Simple Sentences ◽  
Robotic Applications ◽  
Continual Learning

With the recent proliferation of human-oriented robotic applications in domestic and industrial scenarios, it is vital for robots to continually learn about their environments and about the humans they share their environments with. In this paper, we present a novel, online, incremental framework for unsupervised symbol grounding in real-world, human environments for autonomous robots. We demonstrate the flexibility of the framework by learning about colours, people names, usable objects and simple human activities, integrating state-of-the-art object segmentation, pose estimation, activity analysis along with a number of sensory input encodings into a continual learning framework. Natural language is grounded to the learned concepts, enabling the robot to communicate in a human-understandable way. We show, using a challenging real-world dataset of human activities as perceived by a mobile robot, that our framework is able to extract useful concepts, ground natural language descriptions to them, and, as a proof-of-concept, generate simple sentences from templates to describe people and the activities they are engaged in.

scholarly journals Modular Dynamic Neural Network: A Continual Learning Architecture

2021 ◽  
Vol 11 (24) ◽  
pp. 12078
Author(s):  
Daniel Turner ◽  
Pedro J. S. Cardoso ◽  
João M. F. Rodrigues
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Feature Extraction ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Simple Task ◽  
Dynamic Neural Network ◽  
Main Components ◽  
Over Time ◽  
Continual Learning

Learning to recognize a new object after having learned to recognize other objects may be a simple task for a human, but not for machines. The present go-to approaches for teaching a machine to recognize a set of objects are based on the use of deep neural networks (DNN). So, intuitively, the solution for teaching new objects on the fly to a machine should be DNN. The problem is that the trained DNN weights used to classify the initial set of objects are extremely fragile, meaning that any change to those weights can severely damage the capacity to perform the initial recognitions; this phenomenon is known as catastrophic forgetting (CF). This paper presents a new (DNN) continual learning (CL) architecture that can deal with CF, the modular dynamic neural network (MDNN). The presented architecture consists of two main components: (a) the ResNet50-based feature extraction component as the backbone; and (b) the modular dynamic classification component, which consists of multiple sub-networks and progressively builds itself up in a tree-like structure that rearranges itself as it learns over time in such a way that each sub-network can function independently. The main contribution of the paper is a new architecture that is strongly based on its modular dynamic training feature. This modular structure allows for new classes to be added while only altering specific sub-networks in such a way that previously known classes are not forgotten. Tests on the CORe50 dataset showed results above the state of the art for CL architectures.

Does Research-Based Professional Development Make a Difference? A Longitudinal Investigation of Teacher Learning in Technology Integration

2009 ◽  
Vol 111 (5) ◽  
pp. 1195-1241 ◽  
Author(s):  
Chrystalla Mouza
Keyword(s):  
Professional Development ◽  
Teacher Learning ◽  
Professional Growth ◽  
Development Program ◽  
Multiple Sources ◽  
Design Data ◽  
The Impact ◽  
Over Time ◽  
Continual Learning

Background/Context Although there is a growing body of literature on the characteristics of effective professional development, there is little direct evidence on the extent to which these characteristics influence teacher learning and practice. In particular, few studies exist to date that demonstrate the impact of technology-focused professional development on teacher learning and practice. Even fewer studies have examined teacher learning for more than a year to understand the sustainability and growth of professional development gains. Purpose The purpose of this study is to examine the long-term impact of research-based professional development on teacher learning and practice with respect to technology. Analysis is based on data collected from 7 urban teachers 2 years after their participation in a yearlong, technology-focused professional development program. Follow-up data are compared with data collected by the author during the teachers’ participation in professional development to (1) investigate the sustainability and growth of teachers’ learning, (2) identify the conditions that facilitated or hindered teachers’ capacity to further develop their thinking, knowledge, and practice with regard to technology, and (3) map the trajectory of teachers’ learning over a 3-year period. Research Design The study employed a qualitative multiple case study design. Data were collected from multiple sources that included teacher interviews, surveys, classroom observations, and collection of artifacts. Two outcomes were defined as critical measures of long-term learning: sustainability and growth. Findings/Results Results indicated that participation in research-based professional development fostered sustained changes in teachers’ educational technology knowledge, ability to design and implement technology-supported experiences for students, and beliefs toward teaching and learning with technology. In two cases, these changes became the basis for continual learning and led to ongoing professional growth. Further, findings revealed three factors that influenced teacher learning over time: (1) student characteristics, (2) access to resources, and (3) social support and opportunities for collaboration with peers. Conclusions/Recommendations Findings of the study suggest that participation in professional development that is grounded in the currently accepted best practices can impact teacher learning and practice. They also offer insights into the process by which teachers modify their knowledge, practices, and beliefs and the conditions that influence learning over time. Further, they provide new lenses for analyzing teacher learning that suggest looking more closely into the interactive relationship between practices and beliefs, as well as the ways in which classroom experience influences continual learning and change.

Abstract P604: Incidentally-Discovered Asymptomatic Ischemic Lesions Identified Using Natural Language Processing Are Strong Risk Markers for Future Ischemic Stroke

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
David M Kent ◽  
Lester Y Leung ◽  
Yichen Zhou ◽  
Patrick H Luetmer ◽  
David F Kallmes ◽  
...  
Keyword(s):  
Risk Factors ◽  
Ischemic Stroke ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Stroke Risk ◽  
Risk Markers ◽  
Stroke Risk Factors ◽  
Over Time ◽  
Major Stroke

Background: White matter disease (WMD) and silent brain infarction (SBI) are known to be risk markers for stroke. Nevertheless, the predictive value of these changes when seen incidentally on routinely-obtained neuroimages is unknown. Methods: In this retrospective cohort study, Kaiser Permanente-Southern California health plan enrollees aged ≥ 50 years old with a brain CT or MRI scan between 2009-2019 and without a prior history of ischemic stroke, transient ischemic attack, or dementia were identified. Natural language processing (NLP) was used to identify patients with SBI and WMD on the index neuroimaging report. We used Cox proportional hazards to estimate the risk of future ischemic stroke associated with the presence of SBI and of WMD, controlling for major stroke risk factors. Results: Among 262,875 individuals receiving brain neuroimaging, 13,154 (5.0%) and 78,330 (29.8%) had SBI and WMD, respectively. The Table below summarizes the crude stroke incidence rates. The crude hazard ratio (HR) was 3.40 (95% CI 3.25-3.56) for SBI and 2.63 (95% CI 2.54-2.71) for WMD. In the multivariable model controlling for all major stroke risk factors, the effect of SBI was found to be stronger in younger versus older patients and for MRI- versus CT-discovered lesions. With MRI, the average adjusted HR over time was 2.95 (95% CI 2.53-3.44) for those < age 65 and 2.15 (95% CI 1.91-2.41) for those ≥ age 65. With CT scan, the average adjusted HR over time was 2.48 (95%CI 2.19-2.81) for those < age 65 and 1.81 (95% CI 1.71-1.91) for those ≥ age 65. The adjusted HR associated with a finding of WMD was 1.76 (95% CI 1.69-1.82) and was not modified by age or imaging modality. The effect of SBI decreased gradually over time, while the effect of WMD remained constant. Conclusion: Incidentally-discovered SBI and WMD are common in patients ≥ age 50 and are associated with substantial increases in the risk of subsequent symptomatic stroke. The findings may represent an opportunity for stroke prevention.

Transparent Aspect-Level Sentiment Analysis Based on Dependency Syntax Analysis and Its Application on COVID-19

2022 ◽  
Vol 14 (2) ◽  
pp. 1-24
Author(s):  
Bin Wang ◽  
Pengfei Guo ◽  
Xing Wang ◽  
Yongzhong He ◽  
Wei Wang
Keyword(s):  
Data Collection ◽  
Natural Language ◽  
Data Processing ◽  
Sentiment Analysis ◽  
Natural Language Semantic ◽  
Syntax Analysis ◽  
Collection Data ◽  
Dependency Syntax ◽  
Over Time ◽  
Target Words

Aspect-level sentiment analysis identifies fine-grained emotion for target words. There are three major issues in current models of aspect-level sentiment analysis. First, few models consider the natural language semantic characteristics of the texts. Second, many models consider the location characteristics of the target words, but ignore the relationships among the target words and among the overall sentences. Third, many models lack transparency in data collection, data processing, and results generating in sentiment analysis. In order to resolve these issues, we propose an aspect-level sentiment analysis model that combines a bidirectional Long Short-Term Memory (LSTM) network and a Graph Convolutional Network (GCN) based on Dependency syntax analysis (Bi-LSTM-DGCN). Our model integrates the dependency syntax analysis of the texts, and explicitly considers the natural language semantic characteristics of the texts. It further fuses the target words and overall sentences. Extensive experiments are conducted on four benchmark datasets, i.e., Restaurant14, Laptop, Restaurant16, and Twitter. The experimental results demonstrate that our model outperforms other models like Target-Dependent LSTM (TD-LSTM), Attention-based LSTM with Aspect Embedding (ATAE-LSTM), LSTM+SynATT+TarRep and Convolution over a Dependency Tree (CDT). Our model is further applied to aspect-level sentiment analysis on “government” and “lockdown” of 1,658,250 tweets about “#COVID-19” that we collected from March 1, 2020 to July 1, 2020. The experimental results show that Twitter users’ positive and negative sentiments fluctuated over time. Through the transparency analysis in data collection, data processing, and results generating, we discuss the reasons for the evolution of users’ emotions over time based on the tweets and on our models.

Sign in / Sign up

Export Citation Format

Share Document