Carlos Zednik
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Carlos Zednik has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Colin Allen
Department of Philosophy, University of California Santa Barbara 4 , Santa Barbara, CA, USA
1 shared publication - David Landy
Indiana University
1 shared publication - Frank Jäkel
Institute of Cognitive Science, University of Osnabrück, Germany
1 shared publication - Hannes Boelsen
Otto-von-Guericke University, Magdeburg, Germany
1 shared publication
Latest Publications
- Open Access
- 10 March 2022
- Published by Springer Nature in Minds and Machines
- Vol. 32 (1) , 219-239
Scientific Exploration and Explainable Artificial Intelligence
Models developed using machine learning are increasingly prevalent in scientific research. At the same time, these models are notoriously opaque. Explainable AI aims to mitigate the impact of opacity by rendering opaque models transparent. More than being just the solution to a problem, however, Explainable AI can also play an invaluable role in scientific exploration. This paper describes how post-hoc analytic techniques from Explainable AI can be used to refine target phenomena in medical science, to identify starting points for future investigations of (potentially) causal relationships, and to generate possible explanations of target phenomena in cognitive science. In this way, this paper describes how Explainable AI—over and above machine learning itself—contributes to the efficiency and scope of data-driven scientific research.
- 10 March 2022
- Published by Springer Nature in Minds and Machines
- Vol. 32 (1) , 219-239
Open AccessScientific Exploration and Explainable Artificial Intelligence
Models developed using machine learning are increasingly prevalent in scientific research. At the same time, these models are notoriously opaque. Explainable AI aims to mitigate the impact of opacity by rendering opaque models transparent. More than being just the solution to a problem, however, Explainable AI can also play an invaluable role in scientific exploration. This paper describes how post-hoc analytic techniques from Explainable AI can be used to refine target phenomena in medical science, to identify starting points for future investigations of (potentially) causal relationships, and to generate possible explanations of target phenomena in cognitive science. In this way, this paper describes how Explainable AI—over and above machine learning itself—contributes to the efficiency and scope of data-driven scientific research.
- Top Cited Papers
- 20 December 2019
- Published by Springer Nature in Philosophy & Technology
- Vol. 34 (2) , 265-288
Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence
- 20 December 2019
- Published by Springer Nature in Philosophy & Technology
- Vol. 34 (2) , 265-288
Top Cited PapersSolving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence
- Open Access
Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence
Many of the computing systems programmed using Machine Learning are opaque: it is difficult to know why they do what they do or how they work. The Explainable Artificial Intelligence research program aims to develop analytic techniques with which to render opaque computing systems transparent, but lacks a normative framework with which to evaluate these techniques' explanatory success. The aim of the present discussion is to develop such a framework, while paying particular attention to different stakeholders' distinct explanatory requirements. Building on an analysis of 'opacity' from philosophy of science, this framework is modeled after David Marr's influential account of explanation in cognitive science. Thus, the framework distinguishes between the different questions that might be asked about an opaque computing system, and specifies the general way in which these questions should be answered. By applying this normative framework to current techniques such as input heatmapping, feature-detector identification, and diagnostic classification, it will be possible to determine whether and to what extent the Black Box Problem can be solved.
Open AccessSolving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence
Many of the computing systems programmed using Machine Learning are opaque: it is difficult to know why they do what they do or how they work. The Explainable Artificial Intelligence research program aims to develop analytic techniques with which to render opaque computing systems transparent, but lacks a normative framework with which to evaluate these techniques' explanatory success. The aim of the present discussion is to develop such a framework, while paying particular attention to different stakeholders' distinct explanatory requirements. Building on an analysis of 'opacity' from philosophy of science, this framework is modeled after David Marr's influential account of explanation in cognitive science. Thus, the framework distinguishes between the different questions that might be asked about an opaque computing system, and specifies the general way in which these questions should be answered. By applying this normative framework to current techniques such as input heatmapping, feature-detector identification, and diagnostic classification, it will be possible to determine whether and to what extent the Black Box Problem can be solved.
- 13 September 2018
- Published by Taylor & Francis
- p. 357-369
Computational cognitive neuroscience
Computational cognitive neuroscience lies at the intersection of computational neuroscience, which aims to describe structures and processes in the brain through computational modeling and mathematical analysis, and cognitive neuroscience, which aims to explain behavior and cognition through the identification and description of neural mechanisms. This chapter provides a methodological overview of computational cognitive neuroscience, centering on a distinction between two widely used research strategies. On the one hand, top-down strategies are used to infer, from formal characterizations of behavior and cognition, the function and structure of neural mechanisms. On the other hand, bottom-up strategies are used to identify and describe neural mechanisms and their formal properties, and to reconstruct their contributions to specific kinds of behavior and cognition. Many research efforts in computational cognitive neuroscience can be viewed as instances of either the top-down or the bottom-up research strategy.
- 13 September 2018
- Published by Taylor & Francis
- p. 357-369
Computational cognitive neuroscience
Computational cognitive neuroscience lies at the intersection of computational neuroscience, which aims to describe structures and processes in the brain through computational modeling and mathematical analysis, and cognitive neuroscience, which aims to explain behavior and cognition through the identification and description of neural mechanisms. This chapter provides a methodological overview of computational cognitive neuroscience, centering on a distinction between two widely used research strategies. On the one hand, top-down strategies are used to infer, from formal characterizations of behavior and cognition, the function and structure of neural mechanisms. On the other hand, bottom-up strategies are used to identify and describe neural mechanisms and their formal properties, and to reconstruct their contributions to specific kinds of behavior and cognition. Many research efforts in computational cognitive neuroscience can be viewed as instances of either the top-down or the bottom-up research strategy.
- 2 September 2018
- Published by Taylor & Francis in Philosophical Psychology
- Vol. 32 (1) , 23-51
Models and mechanisms in network neuroscience
This paper considers the way mathematical and computational models are used in network neuroscience to deliver mechanistic explanations. Two case studies are considered: Recent work on klinotaxis by Caenorhabditis elegans, and a long-standing research effort on the network basis of schizophrenia in humans. These case studies illustrate the various ways in which network, simulation, and dynamical models contribute to the aim of representing and understanding network mechanisms in the brain, and thus, of delivering mechanistic explanations. After outlining this mechanistic construal of network neuroscience, two concerns are addressed. In response to the concern that functional network models are nonexplanatory, it is argued that functional network models are in fact explanatory mechanism sketches. In response to the concern that models which emphasize a network’s organization over its composition do not explain mechanistically, it is argued that this emphasis is both appropriate and consistent with the principles of mechanistic explanation. What emerges is an improved understanding of the ways in which mathematical and computational models are deployed in network neuroscience, as well as an improved conception of mechanistic explanation in general.
- 2 September 2018
- Published by Taylor & Francis in Philosophical Psychology
- Vol. 32 (1) , 23-51
Models and mechanisms in network neuroscience
This paper considers the way mathematical and computational models are used in network neuroscience to deliver mechanistic explanations. Two case studies are considered: Recent work on klinotaxis by Caenorhabditis elegans, and a long-standing research effort on the network basis of schizophrenia in humans. These case studies illustrate the various ways in which network, simulation, and dynamical models contribute to the aim of representing and understanding network mechanisms in the brain, and thus, of delivering mechanistic explanations. After outlining this mechanistic construal of network neuroscience, two concerns are addressed. In response to the concern that functional network models are nonexplanatory, it is argued that functional network models are in fact explanatory mechanism sketches. In response to the concern that models which emphasize a network’s organization over its composition do not explain mechanistically, it is argued that this emphasis is both appropriate and consistent with the principles of mechanistic explanation. What emerges is an improved understanding of the ways in which mathematical and computational models are deployed in network neuroscience, as well as an improved conception of mechanistic explanation in general.
- 29 August 2018
- Published by Springer Nature
- p. 225-227
Will Machine Learning Yield Machine Intelligence?
- 29 August 2018
- Published by Springer Nature
- p. 225-227
Will Machine Learning Yield Machine Intelligence?
- 6 July 2017
- Published by Taylor & Francis
- p. 389-400
Mechanisms in cognitive science 1
This chapter outlines and defends a mechanistic interpretation of David Marr's account of explanation in cognitive science and thereby attempts to resolve the ambiguities that remain in this account, as well as to extend its scope. It addresses a number of philosophical debates concerning for example the role of idealization in cognitive modelling. The chapter discusses the role of abstraction in mechanistic explanation and the nature of the realization-relationship that obtains between functional processes in the mind and physical structures in the brain. Marr's algorithmic level of analysis is defined by questions about "the representation for the input and output", and about the "algorithm by which the transformation may actually be accomplished". Insofar as researchers across cognitive science answer questions about the what, why, how, and where of behavior and cognition, they are all in the business of mechanistic explanation.
- 6 July 2017
- Published by Taylor & Francis
- p. 389-400
Mechanisms in cognitive science 1
This chapter outlines and defends a mechanistic interpretation of David Marr's account of explanation in cognitive science and thereby attempts to resolve the ambiguities that remain in this account, as well as to extend its scope. It addresses a number of philosophical debates concerning for example the role of idealization in cognitive modelling. The chapter discusses the role of abstraction in mechanistic explanation and the nature of the realization-relationship that obtains between functional processes in the mind and physical structures in the brain. Marr's algorithmic level of analysis is defined by questions about "the representation for the input and output", and about the "algorithm by which the transformation may actually be accomplished". Insofar as researchers across cognitive science answer questions about the what, why, how, and where of behavior and cognition, they are all in the business of mechanistic explanation.
- 5 August 2016
- Published by Springer Nature in Synthese
- Vol. 193 (12) , 3951-3985
Bayesian reverse-engineering considered as a research strategy for cognitive science
- 5 August 2016
- Published by Springer Nature in Synthese
- Vol. 193 (12) , 3951-3985
Bayesian reverse-engineering considered as a research strategy for cognitive science
- 1 January 2015
- Published by Springer Nature
- p. 295-318
Heuristics, Descriptions, and the Scope of Mechanistic Explanation
- 1 January 2015
- Published by Springer Nature
- p. 295-318
Heuristics, Descriptions, and the Scope of Mechanistic Explanation
- Open Access
- 21 April 2014
- Published by Frontiers Media SA in Frontiers in Psychology
- Vol. 5, 275
A perceptual account of symbolic reasoning
People can be taught to manipulate symbols according to formal mathematical and logical rules. Cognitive scientists have traditionally viewed this capacity—the capacity for symbolic reasoning—as grounded in the ability to internally represent numbers, logical relationships, and mathematical rules in an abstract, amodal fashion. We present an alternative view, portraying symbolic reasoning as a special kind of embodied reasoning in which arithmetic and logical formulae, externally represented as notations, serve as targets for powerful perceptual and sensorimotor systems. Although symbolic reasoning often conforms to abstract mathematical principles, it is typically implemented by perceptual and sensorimotor engagement with concrete environmental structures.
- 21 April 2014
- Published by Frontiers Media SA in Frontiers in Psychology
- Vol. 5, 275
Open AccessA perceptual account of symbolic reasoning
People can be taught to manipulate symbols according to formal mathematical and logical rules. Cognitive scientists have traditionally viewed this capacity—the capacity for symbolic reasoning—as grounded in the ability to internally represent numbers, logical relationships, and mathematical rules in an abstract, amodal fashion. We present an alternative view, portraying symbolic reasoning as a special kind of embodied reasoning in which arithmetic and logical formulae, externally represented as notations, serve as targets for powerful perceptual and sensorimotor systems. Although symbolic reasoning often conforms to abstract mathematical principles, it is typically implemented by perceptual and sensorimotor engagement with concrete environmental structures.