A5058670993- Language and cultural evolution
- Natural Language Processing Techniques
- Language, Metaphor, and Cognition
- Categorization, perception, and language
- Hearing Impairment and Communication
- Business Process Modeling and Analysis
- Cognitive Science and Education Research
- Topic Modeling
- Media, Communication, and Education
- Speech and dialogue systems
- Emotion and Mood Recognition
- Service-Oriented Architecture and Web Services
- Animal Vocal Communication and Behavior
- Child and Animal Learning Development
- Semantic Web and Ontologies
- Advanced Database Systems and Queries
- Multisensory perception and integration
- Music and Audio Processing
- Multi-Agent Systems and Negotiation
- Language, Discourse, Communication Strategies
- Authorship Attribution and Profiling
- EEG and Brain-Computer Interfaces
- Hand Gesture Recognition Systems
- Advanced Text Analysis Techniques
- Generative Adversarial Networks and Image Synthesis
University of California, San Diego
2014-2024
Leiden University
2017-2024
Czech Academy of Sciences, Institute of Computer Science
2021
Vrije Universiteit Brussel
2018
Max Planck Society
2017
Max Planck Institute for Psycholinguistics
2017
University of Amsterdam
2009-2014
Delft University of Technology
1987-1991
Developing successful sign language recognition, generation, and translation systems requires expertise in a wide range of fields, including computer vision, graphics, natural processing, human-computer interaction, linguistics, Deaf culture. Despite the need for deep interdisciplinary knowledge, existing research occurs separate disciplinary silos, tackles portions processing pipeline. This leads to three key questions: 1) What does an view current landscape reveal? 2) are biggest...
Developing successful sign language recognition, generation, and translation systems requires expertise in a wide range of fields, including computer vision, graphics, natural processing, human-computer interaction, linguistics, Deaf culture. Despite the need for deep interdisciplinary knowledge, existing research occurs separate disciplinary silos, tackles portions processing pipeline. This leads to three key questions: 1) What does an view current landscape reveal? 2) are biggest...
Abstract In human speech, a finite set of basic sounds is combined into (potentially) unlimited well-formed morphemes. Hockett (1960) placed this phenomenon under the term ‘duality patterning’ and included it as one design features language. Of thirteen proposed, duality patterning least studied still unclear how evolved in Recent work shedding light on summarized paper experimental data presented. This shows that combinatorial structure can emerge an artificial whistled language through...
Abstract Gestures produced by users of spoken languages differ from signs sign in that gestures are more typically ad hoc and idiosyncratic, while conventionalized shared within a language community. To measure how may change over time as result the process conventionalization, we used social coordination game to elicit repeated silent hearing non-signers, Microsoft Kinect unobtrusively track movement their bodies they gestured. Our approach follows tradition laboratory experiments designed...
The impact of introducing double-blind reviewing in the most recent Evolution Language conference is assessed. ranking papers compared between EvoLang 11 (double-blind review) and 9 10 (single-blind review). Main effects were found for first author gender by conference. results mirror some findings literature on review, suggesting that it helps reduce a bias against female authors.
In healthcare, gender and sex considerations are crucial because they affect individuals' health disease differences. Yet, most algorithms deployed in the healthcare context do not consider these aspects account for bias detection. Missing dimensions used medicine is a huge point of concern, as neglecting will inevitably produce far from optimal results generate errors that may lead to misdiagnosis potential discrimination. This paper explores how current algorithmic-based systems reinforce...
This paper reviews how the structure of form and meaning spaces influences nature dynamics form-meaning mappings in language. In general, a structured or space, not all forms meanings are equivalent: some more easily confused with each other than meanings. We first give formalization this idea, explore it robust mappings. It is shown that fundamental properties human language, such as discreteness combinatorial well universals sound systems languages follow from optimal communication spaces....
Abstract In language, recombination of a discrete set meaningless building blocks forms an unlimited possible utterances. How such combinatorial structure emerged in the evolution human language is increasingly being studied. It has been shown that it can emerge when languages culturally evolve and adapt to cognitive biases. emergence interacts with existence holistic iconic form‐meaning mappings still unknown. The experiment presented this paper studies role iconicity learning biases...
Differential Case Marking (DCM) refers to the phenomenon where grammatical case marking is applied selectively based on semantic, pragmatic, or other factors. The emergence of DCM has been studied in artificial language learning experiments with human participants, which were specifically aimed at disentangling effects from those communication (Smith & Culbertson, 2020). Multi-agent reinforcement frameworks neural networks have gained significant interest simulate human-like linguistic...
To understand why music is structured the way it is, we need an explanation that accounts for both universality and variability found in musical traditions. Here test whether statistical universals have been identified melodic structures can emerge as a result of cultural adaptation to human biases through iterated learning. We use data from experiment which artificial whistled systems, where sounds produced with slide whistle were learned by participants transmitted multiple times person...
What is the connection between cultural evolution of a language and rapid processing response to that in brains individual learners? In an iterated communication study was conducted previously, participants were asked communicate temporal concepts such as "tomorrow," "day after," "year," "past" using vertical movements recorded on touch screen. Over time, developed simple artificial 'languages' used space metaphorically nuanced ways about time. Some conventions appeared rapidly universally...
Missing diversity, equity, and inclusion elements in affective computing datasets directly affect the accuracy fairness of emotion recognition algorithms across different groups. A literature review reveals how systems may work differently for groups due to, instance, mental health conditions impacting facial expressions speech or age-related changes appearance health. Our analyzes existing highlights a disconcerting lack diversity current regarding race, sex/gender, age, (mental)...
For the field of education, being able to generate semantically correct and educationally relevant multiple choice questions (MCQs) could have a large impact. While question generation itself is an active research topic, generating distractors (the incorrect options) receives much less attention. A missed opportunity, since there still lot room for improvement in this area. In work, we train GPT-2 language model three given text context, using RACE dataset. Next, BERT answer MCQs, use as...
Abstract. The current generation of CASE (Computer Aided Software Engineering) tools is too inflexible to provide adequate modelling support. One the proposed solutions this problem development so‐called shells. A shell a method independent tool, which may be instantiated with specific become tool supporting that method. As such, provides complete flexibility. This paper does not address benefits shells, as they are completely clear, but focuses on feasibility concept from theoretical well...