Jordi VallverduAutonomous University of Barcelona | UAB · Departament de Filosofia
Jordi Vallverdu
B. Phil, B.Mus, M.Sci, Ph.D
Causality in Deep Learning Research
About
238
Publications
217,476
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,484
Citations
Introduction
ICREA Acadèmia Researcher.
I'm philosopher, with deep interests in epistemology and cognitive sciences, two sides of the same coin. I'm especially devoted to the analysis of the role of emotions within natural and artificial entities. Cultural and natural heuristics are also under my deepest interest. And last but not least, I work on Causality in Deep Learning. Perhaps I'm not a true philosopher, perhaps much better a thinker with several interests. Anyhow, I enjoy doing research!!!
Additional affiliations
January 2002 - March 2017
September 1999 - present
Publications
Publications (238)
Cómo citar este artículo/Citation: Vallverdú, Jordi; Boix, Sarah (2021). Éticas falibles para máquinas (in)falibles. Arbor, 197(800): https://doi.org/10.3989/arbor.2021.800003 Recibido: 21 febrero 2021. Aceptado: 15 mayo 2021. Publicado: RESUMEN: los códigos éticos humanos no son coherentes en su diseño ni tampoco tienen una distribución universal....
Self-adaptive behavior can be defined as the behavior that allows an agent
to adapt to a context using her/his/its resources. The property of being ‘selfadaptive’ implies considering some preliminary sources or elicitors for such
skill. In the case of machine learning, all the learning or self-adaptive behavior mechanisms are related to algorithmic...
Un repaso de lo emocional en los artefactos y los robots
Editorial on the Research Topic
Ethical design of artificial intelligence-based systems for decisionmaking
Background
In this study we investigate the selective compensation of paired peripheral nerves in healthy humans, focusing on distinct axonal conduction velocities in different fibre types. Using paired associative stimulation (PAS) with adjustable parameters, we aimed to modulate and compensate for neuronal activity along the median nerve.
Method...
学术情感学作为情感学（即情感科学）的应用分支，是将情感—情绪神经建模应用于教学过程。本研究首
次深入阐释该学科内涵，并着重论述在自然和人工智能言语行为情感特征分析模型方面的研究成果。鉴于当前对交际
中的情感评估以及行为矩阵中的情绪评估缺乏有效的离散方法，本研究开创性地提出情感智能及其相应的情感人工智
能分级结构，涵盖了总体情感智能、情绪生成智能和情感处理智能。基于仿生概念，我们开发了一种新的“智能助手”模
型———数字讲师，该模型能够赋予教育智能系统情感处理能力，并进一步创建类似于人类的情感人工智能系统。本研究
的实践意义在于，依托人类大脑的神经心理语言学机制，在教学过程中使用神经调节模型，开发深层神经认知构建工具，
并通过生物启发的神经调节架构，为创建具有心理情感状态的计算系统提供了可行性...
The energy efficiency of Artificial Intelligence (AI) systems is a crucial and actual issue that may have an important impact on an ecological, economic and technological level. Spiking Neural Networks (SNNs) are strongly suggested as valid candidates able to overcome Artificial Neural Networks (ANNs) in this specific contest. In this study, the pr...
Este artículo examina el impacto de la inteligencia artificial (IA) y las políticas de vigilancia en la salud menstrual y procreativa desde una perspectiva filosófico-antropológica y ética. La convergencia de la IA y la salud ha generado avances sin precedentes en el diagnóstico, tratamiento y gestión de la salud, pero también plantea desafíos étic...
Este artículo narra la influencia de la Profesora Mercè Izquierdo en la carrera
académica y personal del Prof. Vallverdú, iniciándose en un encuentro fortuito que
se convirtió en una colaboración fructífera. La relación comenzó con el rechazo inicial
del autor hacia la intervención de Izquierdo en su proyecto de tesis, que consideraba
un desvío de...
How can we teach machine learning to identify causal patterns in data? This book explores the very notion of “causality”, identifying from a naturalistic and evolutionary perspective how living systems deal with causal relationships. At the same time, using this knowledge to identify the best ways to apply such biological models in machine learning...
This inaugural chapter provides a comprehensive exploration of causality, delving into its philosophical, symbolic, and scientific dimensions. Beginning with an operational definition using propositional logic ('Α → Β'), the chapter extends this symbolic approach to encompass multicausal scenarios, offering logical representations for diverse causa...
This comprehensive chapter explores the intricate realm of human causal thinking, highlighting its pivotal role in decision-making, problem-solving, and scientific inquiry. The chapter not only emphasizes the strengths of causal thinking but also probes the vulnerabilities inherent in human cognition, including cognitive biases and deviations from...
This chapter delves into the critical need for reliable and interpretable causal algorithms in the contemporary landscape of artificial intelligence. Drawing parallels with Richard III’s desperate cry for a horse, the chapter underscores the pivotal role causal algorithms play in diverse fields, including healthcare, finance, and policy. The challe...
This chapter delves into the intriguing concept of retrocausality, challenging conventional linear causality by proposing that future events could influence the past. Explored in the context of quantum entanglement, delayed-choice experiments, Wheeler–Feynman Absorber Theory, and the transactional interpretation of quantum mechanics, retrocausality...
In the relentless pursuit of advancing artificial intelligence (AI) to emulate human cognition, a crucial frontier lies in achieving a genuine comprehension of causality. This chapter explores the profound significance of endowing AI systems not only with the capacity to recognize causal relationships but also with a true understanding of them. Cau...
This chapter explores the intersection of generative artificial intelligence (AI) and the principles of causality in machine learning, delving into the potential of generative models such as variational autoencoders (VAEs) and generative adversarial networks (GANs) to contribute to causal knowledge. Beyond statistical methodologies, generative AI d...
“Algorithmic causality” encapsulates computational, formal, and epistemic aspects of implementing causal learning into computer systems. This chapter delves into the use of algorithms for understanding and learning causal relationships, which are crucial in artificial intelligence (AI) and machine learning. Drawing from Bishop’s theoretical framewo...
This chapter delves into “Causal AI,” spotlighting achievements, ethics, and future directions. Centered on AI’s evolving aptitude for discerning causal relationships, it explores real-world applications, ethical nuances, and emphasizes robustness, interdisciplinary collaboration, explainability, transparency, and a harmonious human–AI partnership...
Counterfactual thinking in artificial intelligence (AI) is a powerful tool for exploring causal relationships, interpreting decisions, assessing fairness, and evaluating system performance. This chapter delves into the application of counterfactuals across various AI domains such as explainable AI, causal inference, reinforcement learning, fairness...
This chapter delves into the realm of causality in nonhuman minds, exploring how animals demonstrate causal thinking and reasoning. Drawing parallels between machine-like thinking in programming and the nuanced cognition of animals, we examine examples such as Temple Grandin’s empathetic connection with cows and Margaret Hamilton’s systematic appro...
This paper provides a comprehensive exploration of the transformative role of generative AI models, specifically Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), in the realm of medical diagnosis. Drawing from the philosophy of medicine and epidemiology, the paper examines the technical, ethical, and philosophical dimensi...
Inclusive citizen science, an emerging field, has seen extensive research. Prior studies primarily concentrated on creating theoretical models and practical strategies for diversifying citizen science (CS) projects. These studies relied on ethical frameworks or post-project empirical observations. Few examined active participants’ socio-demographic...
En este artículo proponemos una variación del concepto de “máquina antropológica” de Giorgio Agamben en el que las tensiones no se dan entre ser humano y animal, sino entre humano y autómata. Para ilustrarlo, analizamos la representación del autómata en el cine español de los orígenes: El teatro eléctrico de Bob (Chomón, 1909), Muñecos (Perojo, 191...
The "neuropunk revolution" represents an innovative approach focused on the integration of human-machine interfaces through nerve and neural pathways. This cutting-edge concept seeks to push the boundaries of human-machine interaction by exploring the intricate connections between the human nervous system and artificial technologies. Neuropunk delv...
This paper provides a conceptual roadmap for the use of hormonal bioinspired models in a broad range of AI, neuroengineering, or computational systems. The functional signaling nature of hormones provides an example of a reliable multidimensional information management system that can solve parallel multitasks. Two existing examples of hormonal com...
About the planned book:
The relationship between robotics, artificial intelligence (AI), and death raises complex and fundamental issues/questions/problems in philosophy. The fact that we humans live in terms of coexistence with machines with a certain degree of autonomy places us in reality much more shocking than science fiction stories, utopian/...
The unexpended cognitive capacities of plants suggest the possibility of combining them with advances in computation. It is important to explore such a new field of research despite the incompleteness of the empirical support for it.
Why AI does not include gender in its agenda? The role of gender in AI, both as part of the community of agents creating such technologies, as well as part of the contents processed by such technologies is, by far, conflictive. Women have been, again, obliterated by this fundamental revolution of our century. Highly innovative and the first step in...
Parece obvio pensar que los avances neurocientíficos, una revolución de finales del siglo XX, han impactado en la Filosofía Contemporánea. Sin embargo, este no es el caso. Como disciplina compuesta de numerosas áreas, todavía está anclada en modelos antropológicos, culturales, epistemológicos y cognitivos propios de siglos (e incluso milenios) ante...
Fungal organisms can perceive the outer world in a way similar to what animals sense. Does that mean that they have full awareness of their environment and themselves? Is a fungus a conscious entity? In laboratory experiments we found that fungi produce patterns of electrical activity, similar to neurons. There are low and high frequency oscillatio...
In this paper, we provide a brief description of currently existing neural interfaces such as a brain–machine interface, machine–brain interface and bidirectional brain–computer–brain interface. Nevertheless, our aim is not only to provide a review of the current state of the art in these technologies but also to include a philosophical analysis of...
The computer revolution, directly related to the Network society (CASTELLS,
1996, 2010), the Information Age (BYRON, 2010; FLORIDI, 2008), or the 4th Revolution
(SCHWAB, 2017) has provided powerful but also misleading cross-field advances
thanks to Artificial Intelligence. Besides, this related algorithm thinking does not
express the reality and co...
The Research Topic presented here emphasizes the theoretical and methodologicalcontributions of the use of virtual reality to study human cognition. At a first glance, theuse of novel technologies, such as virtual reality can be seen only as broader means ofproviding richer and multimodal stimuli for experimental psychology. However, it should best...
This chapter offers arguments in favor of a morphological characterization of situated abductive processes in perception, considering them as adaptation mechanisms to the varieties of experience. The mechanism that has been analyzed in this essay is the creativity. The thesis defended in this chapter is that the human being maintains a constant hyp...
Toda la ciencia para la sociedad, pero sin la sociedad Toda ciência para a sociedade, mas sem sociedade All Science for Society, but without Society Jordi Vallverdú * Introducción Los estudios CTS no pueden ser enfocados del mismo modo que cualquier otro estudio en el cual el objeto analizado es (relativamente) ajeno al propio sujeto analizador. Po...
New approaches to Uncanny Valley Analysis
This work is dedicated to the review and perspective of the new direction that we call "Neuropunk revolution" resembling the cultural phenomenon of cyberpunk. This new phenomenon has its foundations in advances in neuromorphic technologies including memristive and bio-plausible simulations, BCI, and neurointerfaces as well as unconventional approac...
Today, biomimetic approaches to unconventional computing have been exploring neurotransmitters’ roles. But hormonal aspects of artificial cognition are still unexplored. And from a cognitive perspective, such a complex biochemical combination allows fundamental mechanisms of information processing, from abductive tasks to attention or intentionalit...
Fungal organisms can perceive the outer world in a way similar to what animals sense. Does that mean that they have full awareness of their environment and themselves? Is a fungus a conscious entity? In laboratory experiments we found that fungi produce patterns of electrical activity, similar to neurons. There are low and high frequency oscillatio...
Most of the bioinspired morphological computing studies have departed from a human
analysis bias: to consider cognitive morphology as encapsulated by one body, which, of course, can have enactive connections with other bodies, but that is defined by clear bodily boundaries. Such complex biological inspiration has been directing the research agenda...
El tema de este capítulo, que forma parte el debate sobre el malestar genuino y necesario de los humanos que viven en de-mocracias, se conecta con la naturaleza tanto profundamente emotiva como heurísticamente dispersa de tales agentes. Me he ocupado de ambos elementos en publicaciones anteriores, sin relacionarlos con el debate político relativo a...
In this paper, we suggest that architectures, especially sacred ones, play a significant role to shape cognition. Therefore, cognition is the result of the relationships between the subjects and their surroundings. By sharing the same environment and the relationships with it, the members of a community would weave "common-sense", as a common attit...
Creativity is The Holy Grail of the Cognitive Sciences and it is very important for researchers in the Computer Sciences and AI fields. Although all attempts to explain and replicate intelligence have so far failed, the quest remains a key part of their research. This paper takes two innovative approaches. First, we see cognitive processes as invol...
The use of AI in medical scenarios is currently biased by a conceptual
bias: universal emotions. A real analysis of emotions in real health procedures shows
a completely different perspective: emotional responses in medical practices are
heavily culturally mediated. Therefore, situated and multicultural approaches must
be implemented into medical p...
La creatividad es el santo grial de las ciencias cognitivas y tiene especial importancia para los investigadores de las ciencias de la computación y la inteligencia artificial. Es obvio que la creatividad forma parte esencial de la inteligencia, a pesar de que la mayoría de estudios que intentan explicarla o incluso replicarla han fallado. En el pr...
The arrival of Immersive Virtual and Augmented Reality hardware to the consumer market suggests seamless multi-modal communication between human participants and autonomous interactive characters is an achievable goal in the near future. This possibility is further reinforced by the rapid improvements in the automated analysis of speech, facial exp...
Currently, the main intense debates about the next steps of AI advancements are related to two concepts: the singularity, and AGI (Artificial General Intelligence). By one side, the crucial moment in which AI will reach and surpass the human brain (and collective brain cooperation) power and, by the other, the birth of a conscious and creatively in...
AI has achieved great results using bioinspired ideas. Taking into consideration that information is not a fact to be captured but a process that involves both (even minimal) minds and bodies, the author suggests to use a new two-dimensional or blended approach for the design of AI systems: cultural and natural heuristics, including their “biased”...
En aquest escrit s'ofereix una anàlisi de la situació de pandèmia que estem vivint, aproximant-hi des de la filosofia de la ciència i l'epistemologia. La idea central rau en mostrar que l'epidemiologia, igual que la resta de pràctiques científiques, és una activitat humana i social. Per això, s'ha volgut presentar una dimensió més complerta de la r...
This document provides the framework for the optimum method of applying ECSA’s (European Citizen Science Association) ten principles of citizen science engagement for quality assurance in ethical and inclusive citizen science engagement in the CSI-COP project for the purposes of investigating the extent of online tracking. Following a dynamic recru...
Opinion The first approach to the field of fuckbots, or sexual robotics (also lovotics, dildonics), shows us an astonishing result; we do not have data about them. Taking into consideration that sexual industries (porn, gadgets) provide incredible benefits, the lack of open interest into such research area is, at minimum, a complete absurdity. The...
Public report for EU Horizon2020 CSI-COP project. Grant agreement ID: 873169
This report is a deliverable on the EU Horizon2020 research and innovation project, CSI-COP. It details the literature review conducted to find the best practices in citizen science engagement.
Causality is the most important topic in the history of western science, and since the beginning of the statistical paradigm, its meaning has been reconceptualized many times. Causality entered into the realm of multi‐causal and statistical scenarios some centuries ago. Despite widespread critics, today deep learning and machine learning advances a...
Empathy is a basic emotion trigger for human beings, especially while regulating social relationships and behaviour. The main challenge of this paper is study whether people's empathic reactions towards robots change depending on previous information given to human about the robot before the interaction. The use of false data about robot skills cre...
Human emotions and social processes are evolutionary intertwined, as the result of neuromodulatory mechanisms that define the nature of how bodies interact with the world and create strategies with other bodies and agents. This article presents the previous simulations of TPR, TPR 2.0 and The Game of Emotions. Ideas are also justified in order to a...
With the diffusion and successful new implementation of several machine learning techniques, together with the substantial cost decrease of sensors, also included in mobile devices, the field of emotional analysis and modeling has boosted. Apps, web apps, brain-scanning devices, and Artificial Intelligence assistants often include emotion recogniti...
Since the development of software such as The Logic Theory Machine in 1956 or The General Problem Solver in 1957 the relationship between computers and philosophy has been viewed as a formal, and theoretical one. Computers offered a way to implement and test theoretical philosophical models on language, thought, or methodology.This paper presents e...
Capítulo donde se analizan diversos elementos filosóficos en relación a los videojuegos y la noción de realidad.
Causality is the most important topic in the history of Western Science, and since the beginning of the statistical paradigm, it meaning has been reconceptualized many times. Causality entered into the realm of multi-causal and statistical scenarios some centuries ago. Despite of widespread critics, today Deep Learning and Machine Learning advances...
Spacetime dynamics of crowds are proved to be like dynamics of perturbation waves propagating in an excitable nonlinear medium. This analogy is developed further by studying the propagation of excitation—as represented by the two-variable Oregonator model of the Belousov–Zhabotinsky medium—and by then providing a thorough explanatory application st...
This chapter is a transdisciplinary study on the fundaments of cognition, considering this process at human level as the combination of biological and (multi)cultural values. The main idea is to describe the skill of human beings for using several heuristics for task-solving activities as a process of combining and blending techniques, something th...
In this paper we present a new computational bio-inspired approach. We use the three-dimensional model of emotions created by the Hugo Lövheim “cube of emotions” and validated it via neurosimulation in NEST. We present a computational model that bridges psycho-emotional states with computational processes as the extension of the model “cube of emot...
This edited volume is about how unprejudiced approaches to real human cognition can improve the design of AI. It covers many aspects of human cognition and across 12 chapters the reader can explore multiple approaches about the complexities of human cognitive skills and reasoning, always guided by experts from different but complimentary academic f...
Artificial Intelligence (AI) is the motor that fuels the most profound revolution in the history of humankind. The new era of information society, even called “the Fourth Revolution”, has produced deep changes into sciences, economies, and societies. This revolution is not only about formal heuristics, or the called “algorithmic society”, but also...
This interdisciplinary volume gathers selected, refereed contributions on various aspects of public health from several disciplines and research fields, including the philosophy of science, epidemiology, statistics and ethics. The contributions were originally presented at the 1st Barcelona conference of “Philosophy of Public Health” (5th – 7th May...
Technofeminisms are dealing with new sets of challenges directly related to the possibilities opened by transhumanism. Among the huge list of possible technological advances, we direct our interest towards reproductive technologies, in particular ectogenesis (foetus pregnancy and growth outside a human womb). Although some authors defend the benefi...
The concept of affordance perception is one of the distinctive traits of human cognition; and its application to robots can dramatically improve the quality of human-robot interaction (HRI). In this paper we explore and discuss the idea of "emotional affordances" by proposing a viable model for implementation into HRI; which considers allocentric a...
It has been observed how the technological progress of our civilization mimics the extraordinarily smart design choices of the nature surrounding us. As scientists have been inspired by the cosmos to observe and understand its mechanics from the deepest to the smallest particle, so engineers have found inspiration in biological
systems to design t...
In this paper we present the validation of the three-dimensional model of emotions by Hugo Lövheim the "cube of emotion" via neurosimulation in the NEST. We also present the extension of original "cube of emotion" with the bridge to computational processes parameters. The neurosimulation is done via re-implementation of DA, 5-HT and NA subsystems o...
Affordances are an important concept in cognition, which can be applied to robots in order to perform a successful human-robot interaction (HRI). In this paper we explore and discuss the idea of emotional affordances and propose a viable model for implementation into HRI. We consider “2-ways” affordances: perceived object triggering an emotion, and...
National or intra-national research programs (like FP7 or H2020, in EU zone) foster and reinforce specific ways to do research, selecting the best proposals to implement next researches thanks the provided funds. Very frequently, these programs are based on beliefs about the reliability and best chances of becoming successful researches. We analyse...
This paper summarizes several lexical methods for more comprehensive affect recognition in text using an example of typed utterances. We introduce a set of algorithms that are capable of recognizing emotions of user's statements in order to achieve more effective and smoother human-machine conversation. Aspects often neglected by existing systems w...
Following classic studies on the psychology and cognition of
learning, educators designed teaching heuristics adapted to
the real skills of their students, from early to advanced ages, or even
combining ages in interactive and cooperative active learning by
doing. My point here is to remark the necessity of forcing a Copernican
turn into the educat...
Los problemas éticos se ven delim-tados aquí a tres aspectos: beneficios y perjuicios para los robots; beneficios y perjuicios para los humanos, y bene-ficios y perjuicios para la sociedad. En estos debates hay implícito diversos aspectos que son revisados y analizados en esta reseña.
How Newspeak is a Deep Danger for Science, in this case Epidemiology
The slime mould Physarum polycephalum has been used in developing unconventional computing devices for in which the slime mould played a role of a sensing, actuating, and computing device. These devices treated the slime mould rather as an active living substrate yet the slime mould is a self-consistent living creature which evolved for millions of...
The interest on embodied cognition provides great clues for the understanding of human and other animals data-processing and their social behaviors. On the other side, this understanding offers us new ways to conceptualize and to design from scratch new embodiments for artificial cognitive Systems (AI, robots). We need to take into account that: a)...
Empathy is a basic emotion trigger for human beings, especially while regulating social relationships and behaviour. The main challenge of this paper is study whether people's empathic reactions towards robots change depending on previous information given to human about the robot before the interaction. The use of false data about robot skills cre...
This work aims at demonstrating that the neuromodulatory mechanisms that control the emotional states of mammals (specifically rat’s brains) can be represented and re-implemented in a
computational model processed by a machine. In particular we specifically focus on two neurotransmitters, serotonin and dopamine, starting from their fundamental role...
Si bien la creatividad metodológica es una fuente de avance en las disciplinas científicas, podemos decir que durante su proceso de implementación es al mismo tiempo fuente de confusión. Las controversias científicas producidas a lo largo del siglo XX e inicios del XXI en torno a la epidemiología se nutren o son incentivadas por los debates paralel...
Si bien la creatividad metodológica es una fuente de avance en las disciplinas científicas, podemos decir que durante su proceso de implementación es al mismo tiempo fuente de confusión. Las controversias científicas producidas a lo largo del siglo XX e inicios del XXI en torno a la epidemiología se nutren o son incentivadas por los debates paralel...
The slime mould Physarum polycephalum has been used in developing unconventional computing devices for in which the slime mould played a role of a sensing, actuating, and computing device. These devices treated the slime mould rather as an active living substrate yet the slime mould is a self-consistent living creature which evolved for millions of...
The slime mould Physarum polycephalum has been used in developing un-conventional computing devices for in which the slime mould played a role of a sensing, actuating, and computing device. These devices treated the slime mould rather as an active living substrate yet the slime mould is a self-consistent living creature which evolved for millions o...
One of the challenges of 21 st Century sciences is how to deal with and manage huge amounts of raw data [1] Using several computational tools, scientists are able to capture, process and, finally, to understand that data. The visual aspects of this understanding process are of the utmost importance due to the specific cognitive mechanisms that make...
Human-robot interactions (henceforth HRI) are an increasing research field as well as a real problem for industries and laboratories. The challenges are several, from ethical, economic, cultural or to emotional. New robotic products can or not to be anthropomorphic or adopt ubiquitous shapes and locations, being produced in a range array of forms....
Contemporary scientific approaches to Biology are the result of some cultural ideas considered as universal by Western reductionist traditions. The study of the cultural, symbolic and historical approaches to reality and Life provides us important lessons about the necessity of integrating Eastern holistic views into the study of Life. This is both...
The philosophical differences between Western and Eastern philosophy not only derive from general cultural ideas about reality, but as Nisbet writes (2003), are also methodological, ontological, and cognitively driven. Thus, we can see that strategies of thought and theory-generation are constrained and enabled by conceptual levels, and that the ex...
Questions
Question (1)
Dear colleagues:
for modelization and analysis purposes I'm trying to obtain compared data of number of neurons of both rats and humans at different cerebral levels, such as: Cerebral Cortex, Striatum, Substantia Nigra, Thalamus,....any good source?
Thanks!!!!!
