scholarly journals Explaining Self-Supervised Image Representations with Visual Probing

2021 ◽  
Author(s):  
Dominika Basaj ◽  
Witold Oleszkiewicz ◽  
Igor Sieradzki ◽  
Michał Górszczak ◽  
Barbara Rychalska ◽  
...  
Keyword(s):  
Natural Language ◽  
Language Processing ◽  
Systematic Approach ◽  
Representation Learning ◽  
The Self ◽  
Visual Words ◽  
Semantic Relationships ◽  
Language And Vision ◽  
Supervised Methods ◽  
Image Representations

Recently introduced self-supervised methods for image representation learning provide on par or superior results to their fully supervised competitors, yet the corresponding efforts to explain the self-supervised approaches lag behind. Motivated by this observation, we introduce a novel visual probing framework for explaining the self-supervised models by leveraging probing tasks employed previously in natural language processing. The probing tasks require knowledge about semantic relationships between image parts. Hence, we propose a systematic approach to obtain analogs of natural language in vision, such as visual words, context, and taxonomy. We show the effectiveness and applicability of those analogs in the context of explaining self-supervised representations. Our key findings emphasize that relations between language and vision can serve as an effective yet intuitive tool for discovering how machine learning models work, independently of data modality. Our work opens a plethora of research pathways towards more explainable and transparent AI.

scholarly journals BioVerbNet: a large semantic-syntactic classification of verbs in biomedicine

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Olga Majewska ◽  
Charlotte Collins ◽  
Simon Baker ◽  
Jari Björne ◽  
Susan Windisch Brown ◽  
...  
Keyword(s):  
Natural Language ◽  
Language Processing ◽  
Model Performance ◽  
Representation Learning ◽  
Verb Classes ◽  
Expert Annotation ◽  
Biomedical Texts ◽  
Time Required ◽  
Verb Meaning

Abstract Background Recent advances in representation learning have enabled large strides in natural language understanding; However, verbal reasoning remains a challenge for state-of-the-art systems. External sources of structured, expert-curated verb-related knowledge have been shown to boost model performance in different Natural Language Processing (NLP) tasks where accurate handling of verb meaning and behaviour is critical. The costliness and time required for manual lexicon construction has been a major obstacle to porting the benefits of such resources to NLP in specialised domains, such as biomedicine. To address this issue, we combine a neural classification method with expert annotation to create BioVerbNet. This new resource comprises 693 verbs assigned to 22 top-level and 117 fine-grained semantic-syntactic verb classes. We make this resource available complete with semantic roles and VerbNet-style syntactic frames. Results We demonstrate the utility of the new resource in boosting model performance in document- and sentence-level classification in biomedicine. We apply an established retrofitting method to harness the verb class membership knowledge from BioVerbNet and transform a pretrained word embedding space by pulling together verbs belonging to the same semantic-syntactic class. The BioVerbNet knowledge-aware embeddings surpass the non-specialised baseline by a significant margin on both tasks. Conclusion This work introduces the first large, annotated semantic-syntactic classification of biomedical verbs, providing a detailed account of the annotation process, the key differences in verb behaviour between the general and biomedical domain, and the design choices made to accurately capture the meaning and properties of verbs used in biomedical texts. The demonstrated benefits of leveraging BioVerbNet in text classification suggest the resource could help systems better tackle challenging NLP tasks in biomedicine.

scholarly journals Comparison of Templates with Word2vec in Finding Semantic Relations Between Words

2018 ◽  
pp. 13-17
Author(s):  
Kaan Ant ◽  
Ugur Sogukpinar ◽  
Mehmet Fatif Amasyali
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Large Scale ◽  
Semantic Relations ◽  
Template Method ◽  
Semantic Relationships ◽  
Semantic Spaces

The use of databases those containing semantic relationships between words is becoming increasingly widespread in order to make natural language processing work more effective. Instead of the word-bag approach, the suggested semantic spaces give the distances between words, but they do not express the relation types. In this study, it is shown how semantic spaces can be used to find the type of relationship and it is compared with the template method. According to the results obtained on a very large scale, while is_a and opposite are more successful for semantic spaces for relations, the approach of templates is more successful in the relation types at_location, made_of and non relational.

scholarly journals Essential Features in a Theory of Context for Enabling Artificial General Intelligence

2021 ◽  
Vol 11 (24) ◽  
pp. 11991
Author(s):  
Mayank Kejriwal
Keyword(s):  
Artificial Intelligence ◽  
Natural Language Processing ◽  
Face Recognition ◽  
Natural Language ◽  
Language Processing ◽  
Representation Learning ◽  
Machine Intelligence ◽  
General Intelligence ◽  
Commonsense Reasoning ◽  
Core Concept

Despite recent Artificial Intelligence (AI) advances in narrow task areas such as face recognition and natural language processing, the emergence of general machine intelligence continues to be elusive. Such an AI must overcome several challenges, one of which is the ability to be aware of, and appropriately handle, context. In this article, we argue that context needs to be rigorously treated as a first-class citizen in AI research and discourse for achieving true general machine intelligence. Unfortunately, context is only loosely defined, if at all, within AI research. This article aims to synthesize the myriad pragmatic ways in which context has been used, or implicitly assumed, as a core concept in multiple AI sub-areas, such as representation learning and commonsense reasoning. While not all definitions are equivalent, we systematically identify a set of seven features associated with context in these sub-areas. We argue that such features are necessary for a sufficiently rich theory of context, as applicable to practical domains and applications in AI.

If a picture is worth 1000 words, is a word worth 1000 features for design metric estimation?

2021 ◽  
pp. 1-14
Author(s):  
Kristen Edwards ◽  
Aoran Peng ◽  
Scarlett Miller ◽  
Faez Ahmed
Keyword(s):  
Machine Learning ◽  
Natural Language ◽  
Language Processing ◽  
Design Metrics ◽  
Semantic Relationships ◽  
Consensual Assessment Technique ◽  
Survey Results ◽  
Design Representations ◽  
Expert Ratings ◽  
Do So

Abstract A picture is worth a thousand words, and in design metric estimation, a word may be worth a thousand features. Pictures are awarded this worth because they encode a plethora of information. When evaluating designs, we aim to capture a range of information, including usefulness, uniqueness, and novelty of a design. The subjective nature of these concepts makes their evaluation difficult. Still, many attempts have been made and metrics developed to do so, because design evaluation is integral to the creation of novel solutions. The most common metrics used are the consensual assessment technique (CAT) and the Shah, Vargas-Hernandez, and Smith (SVS) method. While CAT is accurate and often regarded as the “gold standard,” it relies on using expert ratings, making CAT expensive and time-consuming. Comparatively, SVS is less resource-demanding, but often criticized as lacking sensitivity and accuracy. We utilize the complementary strengths of both methods through machine learning. This study investigates the potential of machine learning to predict expert creativity assessments from non-expert survey results. The SVS method results in a text-rich dataset about a design. We utilize these textual design representations and the deep semantic relationships that natural language encodes to predict more desirable design metrics, including CAT metrics. We demonstrate the ability of machine learning models to predict design metrics from the design itself and SVS survey information. We show that incorporating natural language processing improves prediction results across design metrics, and that clear distinctions in the predictability of certain metrics exist.

scholarly journals Designing and Implementing Conversational Intelligent Chat-bot Using Natural Language Processing

2021 ◽  
pp. 262-266
Author(s):  
Asoke Nath ◽  
Rupamita Sarkar ◽  
Swastik Mitra ◽  
Rohitaswa Pradhan
Keyword(s):  
Neural Networks ◽  
Natural Language ◽  
Language Processing ◽  
Deep Neural Networks ◽  
Natural Language Understanding ◽  
Natural Language Generation ◽  
Representation Learning ◽  
Language Understanding ◽  
Language Generation ◽  
User Input

In the early days of Artificial Intelligence, it was observed that tasks which humans consider ‘natural’ and ‘commonplace’, such as Natural Language Understanding, Natural Language Generation and Vision were the most difficult task to carry over to computers. Nevertheless, attempts to crack the proverbial NLP nut were made, initially with methods that fall under ‘Symbolic NLP’. One of the products of this era was ELIZA. At present the most promising forays into the world of NLP are provided by ‘Neural NLP’, which uses Representation Learning and Deep Neural networks to model, understand and generate natural language. In the present paper the authors tried to develop a Conversational Intelligent Chatbot, a program that can chat with a user about any conceivable topic, without having domain-specific knowledge programmed into it. This is a challenging task, as it involves both ‘Natural Language Understanding’ (the task of converting natural language user input into representations that a machine can understand) and subsequently ‘Natural Language Generation’ (the task of generating an appropriate response to the user input in natural language). Several approaches exist for building conversational chatbots. In the present paper, two models have been used and their performance has been compared and contrasted. The first model is purely generative and uses a Transformer-based architecture. The second model is retrieval-based, and uses Deep Neural Networks.

scholarly journals Completing WordNets with Sememe Knowledge

Electronics ◽  
2021 ◽  
Vol 11 (1) ◽  
pp. 79
Author(s):  
Shengwen Li ◽  
Bing Li ◽  
Hong Yao ◽  
Shunping Zhou ◽  
Junjie Zhu ◽  
...  
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Semantic Knowledge ◽  
Graph Structure ◽  
New Words ◽  
Semantic Relationships ◽  
Proposed Model ◽  
Novel Method ◽  
Development And Evolution

WordNets organize words into synonymous word sets, and the connections between words present the semantic relationships between them, which have become an indispensable source for natural language processing (NLP) tasks. With the development and evolution of languages, WordNets need to be constantly updated manually. To address the problem of inadequate word semantic knowledge of “new words”, this study explores a novel method to automatically update the WordNet knowledge base by incorporating word-embedding techniques with sememe knowledge from HowNet. The model first characterizes the relationships among words and sememes with a graph structure and jointly learns the embedding vectors of words and sememes; finally, it synthesizes word similarities to predict concepts (synonym sets) of new words. To examine the performance of the proposed model, a new dataset connected to sememe knowledge and WordNet is constructed. Experimental results show that the proposed model outperforms the existing baseline models.

scholarly journals Distill BERT to Traditional Models in Chinese Machine Reading Comprehension (Student Abstract)

2020 ◽  
Vol 34 (10) ◽  
pp. 13901-13902
Author(s):  
Xingkai Ren ◽  
Ronghua Shi ◽  
Fangfang Li
Keyword(s):  
Reading Comprehension ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Language Model ◽  
Representation Learning ◽  
Original Model ◽  
Language Models ◽  
Machine Reading

Recently, unsupervised representation learning has been extremely successful in the field of natural language processing. More and more pre-trained language models are proposed and achieved the most advanced results especially in machine reading comprehension. However, these proposed pre-trained language models are huge with hundreds of millions of parameters that have to be trained. It is quite time consuming to use them in actual industry. Thus we propose a method that employ a distillation traditional reading comprehension model to simplify the pre-trained language model so that the distillation model has faster reasoning speed and higher inference accuracy in the field of machine reading comprehension. We evaluate our proposed method on the Chinese machine reading comprehension dataset CMRC2018 and greatly improve the accuracy of the original model. To the best of our knowledge, we are the first to propose a method that employ the distillation pre-trained language model in Chinese machine reading comprehension.

scholarly journals A Survey of Reinforcement Learning Informed by Natural Language

2019 ◽  
Author(s):  
Jelena Luketina ◽  
Nantas Nardelli ◽  
Gregory Farquhar ◽  
Jakob Foerster ◽  
Jacob Andreas ◽  
...  
Keyword(s):  
Decision Making ◽  
Reinforcement Learning ◽  
Natural Language ◽  
Language Processing ◽  
Domain Knowledge ◽  
Representation Learning ◽  
World Knowledge ◽  
Language Understanding ◽  
Text Corpora ◽  
The World

To be successful in real-world tasks, Reinforcement Learning (RL) needs to exploit the compositional, relational, and hierarchical structure of the world, and learn to transfer it to the task at hand. Recent advances in representation learning for language make it possible to build models that acquire world knowledge from text corpora and integrate this knowledge into downstream decision making problems. We thus argue that the time is right to investigate a tight integration of natural language understanding into RL in particular. We survey the state of the field, including work on instruction following, text games, and learning from textual domain knowledge. Finally, we call for the development of new environments as well as further investigation into the potential uses of recent Natural Language Processing (NLP) techniques for such tasks.

Sign in / Sign up

Export Citation Format

Share Document