This lesson gives an in-depth introduction of ethics in the field of artificial intelligence, particularly in the context of its impact on humans and public interest. As the healthcare sector becomes increasingly affected by the implementation of ever stronger AI algorithms, this lecture covers key interests which must be protected going forward, including privacy, consent, human autonomy, inclusiveness, and equity.
This lesson describes a definitional framework for fairness and health equity in the age of the algorithm. While acknowledging the impressive capability of machine learning to positively affect health equity, this talk outlines potential (and actual) pitfalls which come with such powerful tools, ultimately making the case for collaborative, interdisciplinary, and transparent science as a way to operationalize fairness in health equity.
This lesson contains both a lecture and a tutorial component. The lecture (0:00-20:03 of YouTube video) discusses both the need for intersectional approaches in healthcare as well as the impact of neglecting intersectionality in patient populations. The lecture is followed by a practical tutorial in both Python and R on how to assess intersectional bias in datasets. Links to relevant code and data are found below.
This lesson describes the fundamentals of genomics, from central dogma to design and implementation of GWAS, to the computation, analysis, and interpretation of polygenic risk scores.
This lesson contains the slides (pptx) of a lecture discussing the necessary concepts and tools for taking into account population stratification and admixture in the context of genome-wide association studies (GWAS). The free-access software Tractor and its advantages in GWAS are also discussed.
This lesson is an overview of transcriptomics, from fundamental concepts of the central dogma and RNA sequencing at the single-cell level, to how genetic expression underlies diversity in cell phenotypes.
This lesson describes the principles underlying functional magnetic resonance imaging (fMRI), diffusion-weighted imaging (DWI), tractography, and parcellation. These tools and concepts are explained in a broader context of neural connectivity and mental health.
This lesson breaks down the principles of Bayesian inference and how it relates to cognitive processes and functions like learning and perception. It is then explained how cognitive models can be built using Bayesian statistics in order to investigate how our brains interface with their environment.
This lesson corresponds to slides 1-64 in the PDF below.
This lecture discusses what defines an integrative approach regarding research and methods, including various study designs and models which are appropriate choices when attempting to bridge data domains; a necessity when whole-person modelling.
This lecture covers different perspectives on the study of the mental, focusing on the difference between Mind and Brain.
This lecture covers a lot of post-war developments in the science of the mind, focusing first on the cognitive revolution, and concluding with living machines.
This lesson briefly goes over the outline of the Neuroscience for Machine Learners course.
This lesson delves into the the structure of one of the brain's most elemental computational units, the neuron, and how said structure influences computational neural network models.
In this lesson you will learn how machine learners and neuroscientists construct abstract computational models based on various neurophysiological signalling properties.
This lesson delves into the human nervous system and the immense cellular, connectomic, and functional sophistication therein.
This lesson describes spike timing-dependent plasticity (STDP), a biological process that adjusts the strength of connections between neurons in the brain, and how one can implement or mimic this process in a computational model. You will also find links for practical exercises at the bottom of this page.
In this lesson, you will learn about some of the many methods to train spiking neural networks (SNNs) with either no attempt to use gradients, or only use gradients in a limited or constrained way.
In this lesson, you will learn how to train spiking neural networks (SNNs) with a surrogate gradient method.
This lesson explores how researchers try to understand neural networks, particularly in the case of observing neural activity.
In this lesson, you will learn about one particular aspect of decision making: reaction times. In other words, how long does it take to take a decision based on a stream of information arriving continuously over time?