Computer scientists have provided a distinct justification of Occam’s Razor. Using the probably approximately correct framework, they provide a theorem that they claim demonstrates that we should favor simpler hypotheses. The argument relies on a philosophical interpretation of the theorem. I argue that the standard interpretation of the result in the literature is misguided and that a better reading does not, in fact, support Occam’s Razor at all. To this end, I state and prove a very similar theorem that, if (...) interpreted the same way, would justify the contradictory Anti-Occam’s Razor—the principle that we should favor more complex hypotheses. (shrink)

We compute that the index set of PAC-learnable concept classes is m-complete Σ30\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\Sigma^{0}_{3}}$$\end{document} within the set of indices for all concept classes of a reasonable form. All concept classes considered are computable enumerations of computable Π10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\Pi^{0}_{1}}$$\end{document} classes, in a sense made precise here. This family of concept classes is sufficient to cover all standard examples, and also has the property that PAC learnability (...) is equivalent to finite VC dimension. (shrink)

The Probably Approximately Correct (PAC) learning is a machine learning model introduced by Leslie Valiant in 1984. The PACi reducibility refers to the PAC reducibility independent of size and computation time. This reducibility in PAC learning resembles the reducibility in Turing computability. The ordering of concept classes under PAC reducibility is nonlinear, even when restricted to particular concrete examples.Due to the resemblance to Turing Reducibility, we suspected that there could be incomparable PACi and PAC degrees for the (...) PACi and PAC reducibilities as in Turing incomparable degrees. In 1957 Friedberg and in 1956 Muchnik independently solved the Post problem by constructing computably enumerable sets A and B of incomparable degrees using the priority construction method. We adapt this idea to PACi and PAC reducibilities and construct two effective concept classes C and D such that C is not reducible to D and vice versa. When considering PAC reducibility it was necessary to work on the size of an effective concept class, thus we use Kolmogorov complexity to obtain the size. The non-learnability of concept classes in the PAC learning model is explained by the existence of PAC incomparable degrees.Analogous to the Turing jump, we give a jump operation on effective concept classes for the zero jump. To define the zero jump operator for PACi degrees the join of all the effective concept classes is constructed and proved that it is a greatest element. There are many properties proven for existing degrees. Thus we can explore proving those properties to PACi and PAC degrees. But if we prove an embedding from those degrees to PACi and PAC degrees then those properties will be true for PACi and PAC degrees without explicitly proving them.Abstract prepared by Dodamgodage Gihnee M. Senadheera and taken directly from the thesisE-mail: [email protected]: https://www.proquest.com/docview/2717762461/abstract/ACD19F29A8774AF6PQ/1?accountid=13864. (shrink)

About 25 years ago, it came to light that a single combinatorial property determines both an important dividing line in model theory and machine learning. The following years saw a fruitful exchange of ideas between PAC-learning and the model theory of NIP structures. In this article, we point out a new and similar connection between model theory and machine learning, this time developing a correspondence between stability and learnability in various settings of online learning. In particular, (...) this gives many new examples of mathematically interesting classes which are learnable in the online setting. (shrink)

This paper starts by summarizing work that philosophers have done in the fields of inductive logic since 1950s and truth approximation since 1970s. It then proceeds to interpret and critically evaluate the studies on machine learning within artificial intelligence since 1980s. Parallels are drawn between identifiability results within formal learning theory and convergence results within Hintikka’s inductive logic. Another comparison is made between the PAC-learning of concepts and the notion of probable approximate truth.

Publication date: 22 December 2020 Source: International Letters of Social and Humanistic Sciences Vol. 89 Author: Oleg Tarnopolsky, Marina Kabanova The article analyzes using Content and Language Integrated Learning for teaching one of the optional humanities disciplines to Ukrainian university students of different majors. The discipline discussed in the article as an example of using CLIL methodology is “The Fundamentals of Psychology and Pedagogy” and it is in the list of optional humanities subjects for the first-year students of Alfred (...) Nobel University in Dnipro, Ukraine. However, unlike the other optional humanities disciplines, the methodology underlying this course is based on teaching the subject in English, instead of Ukrainian, following the CLIL approach widely used in the European education but still little known in Ukraine. The purpose of the paper is to expose and analyze the original methodology developed that conditioned the specific structure of the relevant course, and the specific learning activities used in it. The essence of the developed CLIL theory-based methodology and its practical application are manifested through learning activities that include students’ mini-lectures/workshop-type presentations, brainstorming, case-studies, discussions, and a learning project with its results summarized in students’ essays, abstracts, and summaries written in English. The students collect the information required for completing their tasks not only from the teacher’s lectures and the recommended literature but also by way of doing extensive Internet-search on psychological and pedagogical sites in English. All this makes CLIL in the case under discussion experiential-interactive, blended, autonomous, and cooperative. As the result, students mostly self-construct their knowledge of the discipline by way of using the target language as the tool for such self-construction. This makes the elaborated course a clear-cut case of constructivism in CLIL pedagogy. Subject Classification Numbers: PACS 01.40.-d. (shrink)

Recent years have yielded many discussions on how to endow autonomous agents with the ability to make ethical decisions, and the need for explicit ethical reasoning and transparency is a persistent theme in this literature. We present a modular and transparent approach to equip autonomous agents with the ability to comply with ethical prescriptions, while still enacting pre-learned optimal behaviour. Our approach relies on a normative supervisor module, that integrates a theorem prover for defeasible deontic logic within the control loop (...) of a reinforcement learning agent. The supervisor operates as both an event recorder and an on-the-fly compliance checker w.r.t. an external norm base. We successfully evaluated our approach with several tests using variations of the game Pac-Man, subject to a variety of “ethical” constraints. (shrink)

Objective: Mindfulness-based intervention has been proposed to alleviate loneliness and improve social connectedness. Several randomized controlled trials have been conducted to evaluate the effectiveness of MBI. This study aimed to critically evaluate and determine the effectiveness and safety of MBI in alleviating the feeling of loneliness.Methods: We searched Medline, Embase, PsycInfo, Cochrane CENTRAL, and AMED for publications from inception to May 2020. We included RCTs with human subjects who were enrolled in MBI with loneliness as an outcome. The quality of (...) evidence was assessed using Cochrane's Risk of Bias tool and Grading of Recommendations Assessment, Development, and Evaluation. A random-effects model was used for meta-analysis.Results: Out of 92 articles identified, eight studies involving 815 participants were included in this study. Most trials conducted a minimum of 8 weeks of MBI. Most of the trials used UCLA-Loneliness Scale. A pooled analysis combining three trials and compared with wait-list showed significant improvement in loneliness score reduction using the UCLA-R scale with MD of −6.33 [95% confidence interval : −9.39, −3.26]. Subgroup analysis with only two Cognitively-Based Compassion Training trials also showed similar MD of −6.05. The overall quality of evidence was low.Conclusions: Mindfulness intervention with an average length of 8-week duration significantly improved the population's loneliness level with no mental health issue. However, this evidence had a low GRADE level. (shrink)

In this paper, we perform a worst-case analysis of rule discovery based on generality and accuracy. A rule is defined as a probabilistic constraint of true assignment to the class attribute for corresponding examples. In data mining, a rule can be considered as representing an important class of discovered patterns. We accomplish the aforementioned objective by extending a preliminary version of PAC learning, which represents a worst-case analysis for classification. Our analysis consists of two cases: the case in which (...) we try to avoid finding a bad rule, and the case in which we try to avoid overlooking a good rule. Discussions on related work are also provided for PAC learning, multiple comparison, analysis of association rule discovery, and simultaneous reliability evaluation of a discovered rule. (shrink)

We study computable PAC (CPAC) learning as introduced by Agarwal et al. (2020). First, we consider the main open question of finding characterizations of proper and improper CPAC learning. We give a characterization of a closely related notion of strong CPAC learning, and provide a negative answer to the COLT open problem posed by Agarwal et al. (2021) whether all decidably representable VC classes are improperly CPAC learnable. Second, we consider undecidability of (computable) PAC learnability. We give (...) a simple general argument to exhibit such ndecidability, and initiate a study of the arithmetical complexity of learnability. We briefly discuss the relation to the undecidability result of Ben-David et al. (2019), that motivated the work of Agarwal et al. (shrink)

Indirect negative evidence is clearly an important way for learners to constrain overgeneralisation, and yet a good learning theoretic analysis has yet to be provided for this, whether in a PAC or a probabilistic identification in the limit framework. In this paper we suggest a theoretical analysis of indirect negative evidence that allows the presence of ungrammatical strings in the input and also accounts for the relationship between grammaticality/acceptability and probability. Given independently justified assumptions about lower bounds on the (...) probabilities of grammatical strings, we establish that a limited number of membership queries of some strings can be probabilistically simulated. (shrink)

Humans' voice offers the widest variety of motor phenomena of any human activity. However, its clinical evaluation in people with movement disorders such as Parkinson's disease lags behind current knowledge on advanced analytical automatic speech processing methodology. Here, we use deep learning-based speech processing to differentially analyze voice recordings in 14 people with PD before and after dopaminergic medication using personalized Convolutional Recurrent Neural Networks and Phone Attribute Codebooks. p-CRNN yields an accuracy of 82.35% in the binary classification of (...) ON and OFF motor states at a sensitivity/specificity of 0.86/0.78. The PAC-based approach's accuracy was slightly lower with 73.08% at a sensitivity/specificity of 0.69/0.77, but this method offers easier interpretation and understanding of the computational biomarkers. Both p-CRNN and PAC provide a differentiated view and novel insights into the distinctive components of the speech of persons with PD. Both methods detect voice qualities that are amenable to dopaminergic treatment, including active phonetic and prosodic features. Our findings may pave the way for quantitative measurements of speech in persons with PD. (shrink)

We study model theory of actions of finite groups on substructures of a stable structure. We give an abstract description of existentially closed actions as above in terms of invariants and PAC structures. We show that if the corresponding PAC property is first order, then the theory of such actions has a model companion. Then, we analyze some particular theories of interest (mostly various theories of fields of positive characteristic) and show that in all the cases considered the PAC property (...) is first order. (shrink)

Issues of reliability are claiming center-stage in the epistemology of machine learning. This paper unifies different branches in the literature and points to promising research directions, whilst also providing an accessible introduction to key concepts in statistics and machine learning – as far as they are concerned with reliability.

We introduce and study the notions of a PAC-substructure of a stable structure, and a bounded substructure of an arbitrary substructure, generalizing [10]. We give precise definitions and equivalences, saying what it means for properties such as PAC to be first order, study some examples (such as differentially closed fields) in detail, relate the material to generic automorphisms, and generalize a "descent theorem" for pseudo-algebraically closed fields to the stable context. We also point out that the elementary invariants of pseudo-algebraically (...) closed fields from [6] are also valid for pseudo-differentially closed fields. (shrink)

The essays that follow grew out of a workshop held at the Center for Buddhist Studies, University of California, Berkeley, in March 2018, on the topic of conceptuality and non-conceptuality in Buddhist philosophy. Discussions at the workshop focused specifically on the tenability of the claim made by the two Buddhist epistemologists Dignāga and Dharmakīrti that perceptual cognitions are non-conceptual and yet also contribute to the contents of conceptual thought. The four contributions collected here present just a few of the resulting (...) reflections.This essay explores the Dharmakīrtian1 notion of a "concept" in the specific context of a "pseudo-perception", that is, cognitive events that... (shrink)

We generalize a well-known theorem binding the elementary equivalence relation on the level of PAC fields and the isomorphism type of their absolute Galois groups. Our results concern two cases: saturated PAC structures and nonsaturated PAC structures.

When making decisions, humans can observe many kinds of information about others' activities, but their effects on performance are not well understood. We investigated social learning strategies using a simple problem-solving task in which participants search a complex space, and each can view and imitate others' solutions. Results showed that participants combined multiple sources of information to guide learning, including payoffs of peers' solutions, popularity of solution elements among peers, similarity of peers' solutions to their own, and relative (...) payoffs from individual exploration. Furthermore, performance was positively associated with imitation rates at both the individual and group levels. When peers' payoffs were hidden, popularity and similarity biases reversed, participants searched more broadly and randomly, and both quality and equity of exploration suffered. We conclude that when peers' solutions can be effectively compared, imitation does not simply permit scrounging, but it can also facilitate propagation of good solutions for further cumulative exploration. (shrink)

Learning is fundamentally about action, enabling the successful navigation of a changing and uncertain environment. The experience of pain is central to this process, indicating the need for a change in action so as to mitigate potential threat to bodily integrity. This review considers the application of Bayesian models of learning in pain that inherently accommodate uncertainty and action, which, we shall propose are essential in understanding learning in both acute and persistent cases of pain.

Just as humans can draw conclusions responsibly or irresponsibly, so too can computers. Machine learning systems that have been trained on data sets that include irresponsible judgments are likely to yield irresponsible predictions as outputs. In this paper I focus on a particular kind of inference a computer system might make: identification of the intentions with which a person acted on the basis of photographic evidence. Such inferences are liable to be morally objectionable, because of a way in which (...) they are presumptuous. After elaborating this moral concern, I explore the possibility that carefully procuring the training data for image recognition systems could ensure that the systems avoid the problem. The lesson of this paper extends beyond just the particular case of image recognition systems and the challenge of responsibly identifying a person’s intentions. Reflection on this particular case demonstrates the importance (as well as the difficulty) of evaluating machine learning systems and their training data from the standpoint of moral considerations that are not encompassed by ordinary assessments of predictive accuracy. (shrink)

Perceptual experiences are not immediately responsive to reasons. You see a stick submerged in a glass of water as bent no matter how much you know about light refraction. Due to this isolation from reasons, perception is traditionally considered outside the scope of epistemic evaluability as justified or unjustified. Is perception really as independent from reasons as visual illusions make it out to be? I argue no, drawing on psychological evidence from perceptual learning. The flexibility of perceptual learning (...) is a way of responding to new epistemic reasons. The resulting perceptual experiences are epistemically evaluable as justified or unjustified. (shrink)

This study aimed to identify the learning and business incubation processes and their impact on developing the performance of business incubators in Gaza Strip, and the study relied on the descriptive analytical approach, and the study population consisted of all employees working in business incubators in Gaza Strip in addition to experts and consultants in incubators where their total number reached (62) individuals, and the researchers used the questionnaire as a main tool to collect data through the comprehensive survey (...) method, where (55) questionnaires were retrieved with a recovery rate of (88.7%). The results of the study showed that there is a high approval of the variables of the study, where the dimension of incubation operations came with a relative weight (84.89%), followed by the learning dimension which got a relative weight (82.50%), while the performance level of the incubators got a relative weight (80.12%) and as the study results showed There is a positive relationship statistically between the two dimensions of learning and business incubation processes and performance development in business incubators, and the linear regression analysis model showed that the main effect in improving the performance of business incubators is "after business incubation operations" while it showed weak effect after learning. The study came out with several recommendations, the most important of which is that business incubators seek to reinforce the learning process and incubation processes for business, and the need to periodically evaluate their performance, and that a more effective mechanism be designed to follow up with companies after the end of the incubation period and monitor the progress of these companies. (shrink)

On an account of virtue as skill, virtues are acquired in the ways that skills are acquired. In this paper I focus on one implication of that account that is deserving of greater attention, which is that becoming more skillful requires learning from one’s failures, but that turns out to be especially challenging when dealing with moral failures. In skill acquisition, skills are improved by deliberate practice, where you strive to correct past mistakes and learn how to overcome your (...) current limitations. A similar story applies to virtue acquisition, as moral failures will be a part of anyone’s life, and we will all have to learn from these experiences. However, despite the importance of being able to learn from our mistakes, this is very difficult in practice, given that failure of any kind can be distressing, and especially so for moral failure. The distress created by a recognition of moral failure often prompts responses of anger, avoidance, or defensiveness; rather than attempts to make amends and when necessary to work on self-improvement. The most potentially distressing response to moral failure is shame, as it is often associated with defensiveness. It is here where emotion regulation will be important to manage that distress, and I focus on the skill of emotion differentiation. I argue that emotion differentiation is a promising strategy for distinguishing the emotions we may experience in the wake of failure, including shame, and to encourage those emotions that motivate self-improvement. Thus, emotion regulation is important for virtue acquisition. (shrink)

It is often thought that traditional recidivism prediction tools used in criminal sentencing, though biased in many ways, can straightforwardly avoid one particularly pernicious type of bias: direct racial discrimination. They can avoid this by excluding race from the list of variables employed to predict recidivism. A similar approach could be taken to the design of newer, machine learning-based (ML) tools for predicting recidivism: information about race could be withheld from the ML tool during its training phase, ensuring that (...) the resulting predictive model does not use race as an explicit predictor. However, if race is correlated with measured recidivism in the training data, the ML tool may ‘learn’ a perfect proxy for race. If such a proxy is found, the exclusion of race would do nothing to weaken the correlation between risk (mis)classifications and race. Is this a problem? We argue that, on some explanations of the wrongness of discrimination, it is. On these explanations, the use of an ML tool that perfectly proxies race would (likely) be more wrong than the use of a traditional tool that imperfectly proxies race. Indeed, on some views, use of a perfect proxy for race is plausibly as wrong as explicit racial profiling. We end by drawing out four implications of our arguments. (shrink)

In this chapter, I consider three ‘promises’ of bringing phenomenology into dialogue with networked learning. First, a ‘conceptual promise’, which draws attention to conceptual resources in phenomenology that can inspire and inform how we understand, conceive of, and uncover experiences of participants in networked learning activities and environments. Second, a ‘methodological promise’, which outlines a variety of ways that phenomenological methodologies and concepts can be put to use in empirical research in networked learning. And third, a ‘critical (...) promise’, which suggests how work done in the realm of critical phenomenology is complementary to and useful for exploring the social justice and emancipatory aims of networked learning researchers. Through this framework, I reflect upon multiple ways in which phenomenology has and might continue to fruitfully inform and shape networked learning research. In doing so, I emphasise that deploying phenomenology in the context of networked learning not only enriches our understanding of networked learning but that phenomenology itself is enriched through testing, finetuning, and expanding its methods, concepts, and understanding, as well as unveiling its own limitations and constraints. (shrink)

Testimony is an invaluable source of knowledge. We rely on the reports of those around us for everything from the ingredients in our food and medicine to the identity of our family members. Recent years have seen an explosion of interest in the epistemology of testimony. Despite the multitude of views offered, a single thesis is nearly universally accepted: testimonial knowledge is acquired through the process of transmission from speaker to hearer. In this book, Jennifer Lackey shows that this thesis (...) is false and, hence, that the literature on testimony has been shaped at its core by a view that is fundamentally misguided. She then defends a detailed alternative to this conception of testimony: whereas the views currently dominant focus on the epistemic status of what speakers believe, Lackey advances a theory that instead centers on what speakers say. The upshot is that, strictly speaking, we do not learn from one another's beliefs - we learn from one another's words. Once this shift in focus is in place, Lackey goes on to argue that, though positive reasons are necessary for testimonial knowledge, testimony itself is an irreducible epistemic source. This leads to the development of a theory that gives proper credence to testimony's epistemologically dual nature: both the speaker and the hearer must make a positive epistemic contribution to testimonial knowledge. The resulting view not only reveals that testimony has the capacity to generate knowledge, but it also gives appropriate weight to our nature as both socially indebted and individually rational creatures. The approach found in this book will, then, represent a radical departure from the views currently dominating the epistemology of testimony, and thus is intended to reshape our understanding of the deep and ubiquitous reliance we have on the testimony of those around us. (shrink)

According to certain normative theories in epistemology, rationality requires us to be logically omniscient. Yet this prescription clashes with our ordinary judgments of rationality. How should we resolve this tension? In this paper, I focus particularly on the logical omniscience requirement in Bayesian epistemology. Building on a key insight by Hacking :311–325, 1967), I develop a version of Bayesianism that permits logical ignorance. This includes: an account of the synchronic norms that govern a logically ignorant individual at any given time; (...) an account of how we reduce our logical ignorance by learning logical facts and how we should update our credences in response to such evidence; and an account of when logical ignorance is irrational and when it isn’t. At the end, I explain why the requirement of logical omniscience remains true of ideal agents with no computational, processing, or storage limitations. (shrink)

Explores the resources for contemporary ethics found in the work of the Cheng brothers, canonical neo-Confucian philosophers. Yong Huang presents a new way of doing comparative philosophy as he demonstrates the resources for contemporary ethics offered by the Cheng brothers, Cheng Hao (1032–1085) and Cheng Yi (1033–1107), canonical neo-Confucian philosophers. Huang departs from the standard method of Chinese/Western comparison, which tends to interest those already interested in Chinese philosophy. While Western-oriented scholars may be excited to learn about Chinese philosophers who (...) have said things similar to what they or their favored philosophers have to say, they hardly find anything philosophically new from such comparative work. Instead of comparing and contrasting philosophers, each chapter of this book discusses a significant topic in Western moral philosophy, examines the representative views on this topic in the Western tradition, identifies their respective difficulties, and discusses how the Cheng brothers have better things to say on the subject. Topics discussed include why one should be moral, how weakness of will is not possible, whether virtue ethics is self-centered, in what sense the political is also personal, how a moral theory can be of an antitheoretical nature, and whether moral metaphysics is still possible in this postmodern and postmetaphysical age. “This book presents the philosophical ideas of the Cheng brothers intelligently, convincingly, and powerfully. It is among the best books ever written on the Cheng brothers, including works in the Chinese language.” — Kam-por Yu, coeditor of Taking Confucian Ethics Seriously: Contemporary Theories and Applications. (shrink)

Discusses the role of education in liberating the oppressed people of the Third World.

In this article, we explore the learning of newcomers in a religious community through a micro-sociological approach, making use of Jean Lave and Etienne Wenger’s (1991) notion of “legitimate peripheral participation” to conceptualize initial stages of inclusion and involvement in social practice. Our case study concerns Orthodox Christianity and is based on material gathered through fieldwork in a course targeting potential new members organized by a Finnish Orthodox parish. In the analysis, we inquire into how beginners learn skilful participation (...) in Orthodox liturgical life, and specifically embodied ritual conduct. This learning takes place primarily through participation in real-life divine services. The article highlights challenges faced by beginners in acquiring the embodied repertoire of Orthodox ritual, including adapting to the artistic use of ritual gestures, and negotiating the meanings produced through them. Furthermore, it also illustrates how nuanced dynamics between newcomers and old-timers influence the learning process. (shrink)

This is an excerpt of a report that highlights and explores five questions that arose from the Network for Sensory Research workshop on perceptual learning and perceptual recognition at the University of York in March, 2012. This portion of the report explores the question: What is perceptual learning?

In this new volume in the Oxford Psychology Series, the author presents a highly readable account of the cognitive unconscious, focusing in particular on the problem of implicit learning. Implicit learning is defined as the acquisition of knowledge that takes place independently of the conscious attempts to learn and largely in the absence of explicit knowledge about what was acquired. One of the core assumptions of this argument is that implicit learning is a fundamental, "root" process, one (...) that lies at the very heart of the adaptive behavioral repertoire of every complex organism. The author's goals are to outline the essential features of implicit learning that have emerged from the many studies that have been carried out in a variety of experimental laboratories over the past several decades; to present the various alternative perspectives on this issue that have been proposed by other researchers and to try to accommodate these views with his own; to structure the literature so that it can be seen in the context of standard heuristics of evolutionary biology; to present the material within a functionalist approach and to try to show why the experimental data should be seen as entailing particular epistemological perspectives; and to present implicit processing as encompassing a general and ubiquitous set of operations that have wide currency and several possible applications. Chapter 1 begins with the core problem under consideration in this book, a characterization of "implicit learning" as it has come to be used in the literature. Reber puts this seemingly specialized topic into a general framework and suggests a theoretical model based on standard heuristics of evolutionary biology. In his account, Reber weaves a capsule history of interest in and work on the cognitive unconscious. Chapter 2 turns to a detailed overview of the experimental work on the acquisition of implicit knowledge, which currently is of great interest. Chapter 3 develops the evolutionary model within which one can see learning and cognition as richly intertwining issues and not as two distinct fields with one dominating the other. Finally, Chapter 4 explores a variety of entailments and speculations concerning implicit cognitive processes and their general role in the larger scope of human performance. (shrink)