Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
121
result(s) for
"Carrasco, Xavier"
Sort by:
Rumpelstiltskin
A strange little man helps the miller's daughter spin straw into gold for the king on the condition that she will give him her first-born child.
Book
Human–machine co-creation: a complementary cognitive approach to creative character design process using GANs
Recent advances in generative adversarial networks (GANs) applications continue to attract the attention of researchers in different fields. In such a framework, two neural networks compete adversely to generate new visual contents indistinguishable from the original dataset. The objective of this research is to create a complementary co-design process between humans and machines to augment character designers’ abilities in visualizing and creating new characters for multimedia projects such as games and animation. Driven by design cognitive scaffolding, the proposed approach aims to inform the processes of
perceiving
,
knowing
, and
making
. The machine-generated concepts are used as a launching platform for character designers to conceptualize new characters. A labelled dataset of 22
,
000 characters was developed for this work and deployed using different GANs to evaluate the most suited for the context, followed by mixed methods evaluation for the machine output and human derivations. The discussed results substantiate the value of the proposed co-creation framework and elucidate how the generated concepts are used as cognitive substances that interact with designers’ competencies in a versatile manner to influence the creative processes of conceptualizing novel characters.
Journal Article
Augmenting Character Designers Creativity Using Generative Adversarial Networks
Recent advances in Generative Adversarial Networks (GANs) continue to attract the attention of researchers in different fields due to the wide range of applications devised to take advantage of their key features. Most recent GANs are focused on realism, however, generating hyper-realistic output is not a priority for some domains, as in the case of this work. The generated outcomes are used here as cognitive components to augment character designers creativity while conceptualizing new characters for different multimedia projects. To select the best-suited GANs for such a creative context, we first present a comparison between different GAN architectures and their performance when trained from scratch on a new visual characters dataset using a single Graphics Processing Unit. We also explore alternative techniques, such as transfer learning and data augmentation, to overcome computational resource limitations, a challenge faced by many researchers in the domain. Additionally, mixed methods are used to evaluate the cognitive value of the generated visuals on character designers agency conceptualizing new characters. The results discussed proved highly effective for this context, as demonstrated by early adaptations to the characters design process. As an extension for this work, the presented approach will be further evaluated as a novel co-design process between humans and machines to investigate where and how the generated concepts are interacting with and influencing the design process outcome.
Paper
Entering the Era of Discrete Diffusion Models: A Benchmark for Schrödinger Bridges and Entropic Optimal Transport
The Entropic Optimal Transport (EOT) problem and its dynamic counterpart, the Schrödinger bridge (SB) problem, play an important role in modern machine learning, linking generative modeling with optimal transport theory. While recent advances in discrete diffusion and flow models have sparked growing interest in applying SB methods to discrete domains, there remains no reliable way to assess how well these methods actually solve the underlying problem. We address this challenge by introducing a benchmark for SB on discrete spaces. Our construction yields pairs of probability distributions with analytically known SB solutions, enabling rigorous evaluation. As a byproduct of building this benchmark, we obtain two new SB algorithms, DLightSB and DLightSB-M, and additionally extend prior related work to construct the \\(\\)-CSBM algorithm. We demonstrate the utility of our benchmark by evaluating both existing and new solvers in high-dimensional discrete settings. This work provides the first step toward proper evaluation of SB methods on discrete spaces, paving the way for more reproducible future studies. The code for the benchmark and all associated experiments is available at https://github.com/gregkseno/catsbench.
Paper
Uncovering Challenges of Solving the Continuous Gromov-Wasserstein Problem
Recently, the Gromov-Wasserstein Optimal Transport (GWOT) problem has attracted the special attention of the ML community. In this problem, given two distributions supported on two (possibly different) spaces, one has to find the most isometric map between them. In the discrete variant of GWOT, the task is to learn an assignment between given discrete sets of points. In the more advanced continuous formulation, one aims at recovering a parametric mapping between unknown continuous distributions based on i.i.d. samples derived from them. The clear geometrical intuition behind the GWOT makes it a natural choice for several practical use cases, giving rise to a number of proposed solvers. Some of them claim to solve the continuous version of the problem. At the same time, GWOT is notoriously hard, both theoretically and numerically. Moreover, all existing continuous GWOT solvers still heavily rely on discrete techniques. Natural questions arise: to what extent do existing methods unravel the GWOT problem, what difficulties do they encounter, and under which conditions they are successful? Our benchmark paper is an attempt to answer these questions. We specifically focus on the continuous GWOT as the most interesting and debatable setup. We crash-test existing continuous GWOT approaches on different scenarios, carefully record and analyze the obtained results, and identify issues. Our findings experimentally testify that the scientific community is still missing a reliable continuous GWOT solver, which necessitates further research efforts. As the first step in this direction, we propose a new continuous GWOT method which does not rely on discrete techniques and partially solves some of the problems of the competitors.
Paper
Entering the Era of Discrete Diffusion Models: A Benchmark for Schrödinger Bridges and Entropic Optimal Transport
The Entropic Optimal Transport (EOT) problem and its dynamic counterpart, the Schrödinger bridge (SB) problem, play an important role in modern machine learning, linking generative modeling with optimal transport theory. While recent advances in discrete diffusion and flow models have sparked growing interest in applying SB methods to discrete domains, there is still no reliable way to evaluate how well these methods actually solve the underlying problem. We address this challenge by introducing a benchmark for SB on discrete spaces. Our construction yields pairs of probability distributions with analytically known SB solutions, enabling rigorous evaluation. As a byproduct of building this benchmark, we obtain two new SB algorithms, DLightSB and DLightSB-M, and additionally extend prior related work to construct the \\(\\)-CSBM algorithm. We demonstrate the utility of our benchmark by evaluating both existing and new solvers in high-dimensional discrete settings. This work provides the first step toward proper evaluation of SB methods on discrete spaces, paving the way for more reproducible future studies.
Paper
Uncovering Challenges of Solving the Continuous Gromov-Wasserstein Problem
Recently, the Gromov-Wasserstein Optimal Transport (GWOT) problem has attracted the special attention of the ML community. In this problem, given two distributions supported on two (possibly different) spaces, one has to find the most isometric map between them. In the discrete variant of GWOT, the task is to learn an assignment between given discrete sets of points. In the more advanced continuous formulation, one aims at recovering a parametric mapping between unknown continuous distributions based on i.i.d. samples derived from them. The clear geometrical intuition behind the GWOT makes it a natural choice for several practical use cases, giving rise to a number of proposed solvers. Some of them claim to solve the continuous version of the problem. At the same time, GWOT is notoriously hard, both theoretically and numerically. Moreover, all existing continuous GWOT solvers still heavily rely on discrete techniques. Natural questions arise: to what extent existing methods unravel GWOT problem, what difficulties they encounter, and under which conditions they are successful. Our benchmark paper is an attempt to answer these questions. We specifically focus on the continuous GWOT as the most interesting and debatable setup. We crash-test existing continuous GWOT approaches on different scenarios, carefully record and analyze the obtained results, and identify issues. Our findings experimentally testify that the scientific community is still missing a reliable continuous GWOT solver, which necessitates further research efforts. As the first step in this direction, we propose a new continuous GWOT method which does not rely on discrete techniques and partially solves some of the problems of the competitors. Our code is available at https://github.com/Ark-130994/GW-Solvers.
Paper
Weighting health-related estimates in the GCAT cohort and the general population of Catalonia
Population-based cohorts play a key role in personalized medicine. However, it is known that cohorts are affected by the “healthy volunteer bias” where participants are generally healthier than the broader population, compromising its representativeness. Here, we assess the healthy bias, identifying bias key indicators for representativeness of the GCAT cohort, encompassing 20,000 adult participants of Catalonia, and generating survey raked weights to enhance the cohort’s comparability. To assess and correct the bias, we compare multiple variables across sociodemographic, lifestyle, diseases and medication domains. Electronic health records of Catalonia (SIDIAP), the Health Survey of Catalonia (ESCA) and registers from the statistics institute of Catalonia (IDESCAT) and Spain (INE) were used to make the comparisons. We observed that the GCAT cohort is enriched in women and younger individuals, people with higher socioeconomic status and more health conscious and healthier individuals in terms of mortality and chronic disease prevalence. Raked survey weighting identified sex, birth year, rurality, education level, civil status, occupation status, smoking habit, household size, self-perceived health status and number of primary care visits as key weight variables. On average, raked weights reduced the differences by 70% for compared variables, and by 26% in disease prevalence estimates. We can conclude that the application of raked weights has enhanced the cohort’s representativeness, improved comparability, and yielded more precise estimates when analysing GCAT data.
Journal Article
The role of satisfaction with the way you look in 10 to 12-year-olds’ health and subjective well-being across genders and cultures
Few studies have examined the relationship between health and well-being in the general population of children and adolescents from a psychosocial perspective. The present article therefore addresses the following unprecedented twofold objective: (1) to conduct an in-depth analysis of differences in satisfaction with the way you look, satisfaction with health and a global measure of subjective well-being (SWB), by age group (10- and 12-year-olds), gender and country; and (2) to explore to what extent satisfaction with the way you look mediates the relationship between satisfaction with health and a global measure of SWB, and how the relationship between these variables is related to gender and country of origin in the two mentioned age groups. Using a cross-sectional and correlational approach, analyses were conducted using data obtained from a self-administered questionnaire on a sample of 91,076 adolescents from 35 countries belonging to the Children’s Worlds 3rd wave dataset. Satisfaction with the way you look was found to mediate the relationship between satisfaction with health and a global measure of SWB, and the variables age, gender and country were identified as being important in this mediation. These findings further our knowledge on the contribution of SWB-related evaluations in these age groups for different countries and genders, but also on the adoption of a broader and more preventive approach to improving them. The relevance of implementing interventions focused on improving self-perception at these ages is discussed.
Journal Article
A methodological review finds mismatch between overall and pairwise overlap analysis in a sample of overviews
Overlap of primary studies is a key methodological challenge for overviews. There are limited reports of methods used to address overlap, and there is no detailed assessment of the corrected covered area (CCA) of a representative sample of overviews. To describe the approaches used to address overlap, and to estimate the overall and pairwise CCA.
We searched PubMed for overviews published in 2018. Two authors conducted the screening process. We described the strategy used for assessing overlap, and calculated overall and pairwise CCA for each overview.
We analyzed a random sample of 30 out of 89 eligible articles. Eleven did not address the overlap. Of the remainder, most frequent strategies were visual assessment and discussion of overlap as a limitation. Median overall CCA among the included overviews was 6.7%. The pairwise analysis showed that 52.8% of SR pairs had slight overlap, while 28.3% had very high overlap.
Reported strategies for addressing overlap vary considerably among overview authors. The pairwise approach for assessing the CCA revealed highly overlapped pairs of SRs in overviews with overall slight overlap and vice versa. We encourage authors to complement the overall CCA assessment with a pairwise approach.
•The overlap of primary studies is a key challenge for overviews.•Nearly a third of authors do not report a strategy to handle overlap.•The corrected covered area formula is an accepted approach to measure overlap.•Our findings show an overall moderate overlap degree.•One-third of the overviews showed mismatch between the overall and pairwise overlap.
Journal Article