The purpose of this project is to argue that we possess a minimal self. It will demonstrate that minimal selfhood arrives early in our development and continues to remain and influence us throughout our entire life. There are two areas of research which shape my understanding of the minimal self: phenomenology and enactivism. Phenomenology emphasizes the sense of givenness, ownership, or mineness that accompanies all of our experiences. Enactivism says there is a sensorimotor coupling that occurs between us and the environment in a way which modulates the dynamic patterns of our self development; the laying down of these basic patterns helps make us who we are and gives rise to the phenomenological, experiential mineness. Drawing on these two core ideas, I will be arguing for a Phenomenological-Enactive Minimal Self (abbreviated PEMS). I will be emphasizing the role of the body and the role of affects (moods, feelings, and emotions) as the most important components relevant to understanding minimal selfhood. Put more concretely, the set of conditions which constitute the PEMS view are: (i) The minimal self is the experiential subject; the minimal sense of self is present whenever there is awareness. It is the subjectivity of experience, the sense of mineness, or givenness which our experiences contain. (ii) The phenomenological part of the PEMS view turns on the idea of a bodily and dynamic integration of sensorimotor coupling and affective experience. It is, ontologically speaking, the lived body in enactive engagement with the environment. It is this embodied subject which anchors and forms the foundation for the later ‘narrative' self, which emerges from it and which is continually influenced by it. It is the subject enactively engaged with others, dependent on sensorimotor processes and affects. We have an identity, but it emerges from relational and dynamic processes.

There is a conceptual knot at the intersection of moral responsibility and action theory. This knot can be expressed as the following question: What is the relationship between an agent's openness to moral responsibility and the intentional status of her behaviour? My answer to this question is developed in three steps. I first develop a control-backed account of intentional agency, one that borrows vital insights from the cognitive sciences – in the form of Dual Process Theory – in understanding the control condition central to the account, and demonstrate that this account fares at least as well as its rivals in the field. Secondly, I investigate the dominant positions in the discussion surrounding the role of control in moral responsibility. After consideration of some shortcomings of these positions – especially the inability to properly account for so-called ambivalence cases – I defend an alternative pluralist account of moral responsibility, in which there are two co-extant variants of such responsibility: attributability and accountability. The latter of these will be shown to have a necessary control condition, also best understood in terms of a requirement for oversight (rather than conscious or online control), and in terms of the workings of the dual system mechanism. I then demonstrate how these two accounts are necessarily related through the shared role of this kind of control, leading to my answer to the original question: if an agent is open to moral accountability based on some activity or outcome, this activity or outcome must necessarily have positive intentional status. I then apply this answer in a consideration of certain cases of the use of the Doctrine of Double Effect.

La théorie du processus dual stipule que la cognition humaine fonctionne selon deux modes distincts : l’un pour le traitement rapide, habituel et associatif, appelé communément "système 1" et le second, ayant un traitement plus lent, délibéré et contrôlé, que l’on nomme "système 2". Cette distinction indique une caractéristique sous-jacente importante de la cognition humaine : la possibilité de passer de manière adaptative à différentes stratégies de calcul selon la situation. Cette capacité est étudiée depuis longtemps dans différents domaines et de nombreux bénéfices hypothétiques semblent y être liés. Cependant, les réseaux neuronaux profonds sont souvent construits sans cette capacité à gérer leurs ressources calculatoires de manière optimale. Cette limitation des modèles actuels est d’autant plus préoccupante que de plus en plus de travaux récents semblent montrer une relation linéaire entre la capacité de calcul utilisé et les performances du modèle lors de la phase d’évaluation. Pour résoudre ce problème, ce mémoire propose différentes approches et étudie leurs impacts sur les modèles, tout d’abord, nous étudions un agent d’apprentissage par renforcement profond qui est capable d’allouer plus de calcul aux situations plus difficiles. Notre approche permet à l’agent d’adapter ses ressources computationnelles en fonction des exigences de la situation dans laquelle il se trouve, ce qui permet en plus d’améliorer le temps de calcul, améliore le transfert entre des tâches connexes et la capacité de généralisation. L’idée centrale commune à toutes nos approches est basée sur les théories du coût de l’effort venant de la littérature sur le contrôle cognitif qui stipule qu’en rendant l’utilisation de ressource cognitive couteuse pour l’agent et en lui laissant la possibilité de les allouer lors de ses décisions il va lui-même apprendre à déployer sa capacité de calcul de façon optimale. Ensuite, nous étudions des variations de la méthode sur une tâche référence d’apprentissage profond afin d’analyser précisément le comportement du modèle et quels sont précisément les bénéfices d’adopter une telle approche. Nous créons aussi notre propre tâche "Stroop MNIST" inspiré par le test de Stroop utilisé en psychologie afin de valider certaines hypothèses sur le comportement des réseaux neuronaux employant notre méthode. Nous finissons par mettre en lumière les liens forts qui existent entre apprentissage dual et les méthodes de distillation des connaissances. Notre approche a la particularité d’économiser des ressources computationnelles lors de la phase d’inférence. Enfin, dans la partie finale, nous concluons en mettant en lumière les contributions du mémoire, nous détaillons aussi des travaux futurs, nous approchons le problème avec les modèles basés sur l’énergie, en apprenant un paysage d’énergie lors de l’entrainement, le modèle peut ensuite lors de l’inférence employer une capacité de calcul dépendant de la difficulté de l’exemple auquel il fait face plutôt qu’une simple propagation avant fixe ayant systématiquement le même coût calculatoire. Bien qu’ayant eu des résultats expérimentaux infructueux, nous analysons les promesses que peuvent tenir une telle approche et nous émettons des hypothèses sur les améliorations potentielles à effectuer. Nous espérons, avec nos contributions, ouvrir la voie vers des algorithmes faisant un meilleur usage de leurs ressources computationnelles et devenant par conséquent plus efficace en termes de coût et de performance, ainsi que permettre une compréhension plus intime des liens qui existent entre certaines méthodes en apprentissage machine et la théorie du processus dual. / The dual-process theory states that human cognition operates in two distinct modes: one for rapid, habitual and associative processing, commonly referred to as "system 1", and the second, with slower, deliberate and controlled processing, which we call "system 2". This distinction points to an important underlying feature of human cognition: the ability to switch adaptively to different computational strategies depending on the situation. This ability has long been studied in various fields, and many hypothetical benefits seem to be linked to it. However, deep neural networks are often built without this ability to optimally manage their computational resources. This limitation of current models is all the more worrying as more and more recent work seems to show a linear relationship between the computational capacity used and model performance during the evaluation phase. To solve this problem, this thesis proposes different approaches and studies their impact on models. First, we study a deep reinforcement learning agent that is able to allocate more computation to more difficult situations. Our approach allows the agent to adapt its computational resources according to the demands of the situation in which it finds itself, which in addition to improving computation time, enhances transfer between related tasks and generalization capacity. The central idea common to all our approaches is based on cost-of-effort theories from the cognitive control literature, which stipulate that by making the use of cognitive resources costly for the agent, and allowing it to allocate them when making decisions, it will itself learn to deploy its computational capacity optimally. We then study variations of the method on a reference deep learning task, to analyze precisely how the model behaves and what the benefits of adopting such an approach are. We also create our own task "Stroop MNIST" inspired by the Stroop test used in psychology to validate certain hypotheses about the behavior of neural networks employing our method. We end by highlighting the strong links between dual learning and knowledge distillation methods. Finally, we approach the problem with energy-based models, by learning an energy landscape during training, the model can then during inference employ a computational capacity dependent on the difficulty of the example it is dealing with rather than a simple fixed forward propagation having systematically the same computational cost. Despite unsuccessful experimental results, we analyze the promise of such an approach and speculate on potential improvements. With our contributions, we hope to pave the way for algorithms that make better use of their computational resources, and thus become more efficient in terms of cost and performance, as well as providing a more intimate understanding of the links that exist between certain machine learning methods and dual process theory.