Online research in philosophy

Two great problems of learning confront humanity: learning about the universe, and learning how to live wisely. The first problem was solved with the creation of modern science, but the second problem has not been solved. This combination puts humanity into a situation of unprecedented danger. In order to solve the second problem we need to learn from our solution to the first problem. This requires that we bring about a revolution in the overall aims and methods of academic inquiry, so that it takes up its proper task of promoting wisdom.

Sequential behavior is essential to intelligence, and it is a fundamental part of human activities ranging from reasoning to language, and from everyday skills to complex problem solving. In particular, sequence learning is an important component of learning in many task domains — planning, reasoning, robotics, natural language processing, speech recognition, adaptive control, time series prediction, financial engineering, DNA sequencing, and so on. Naturally, there are many different approaches towards sequence learning, resulting from different perspectives taken in different task domains. These approaches deal with somewhat differently formulated sequential learning problems (for example, some with actions and some without), and/ or different aspects of sequence learning (for example, sequence prediction vs. sequence recognition). Sequence learning is clearly a difiicult task. More powerful algorithms for sequence learning are needed in all of these afore-mentioned domains. It is our view that the right approach to develop better techniques, algorithms, models.

Given the privileged status claimed for active learning in a variety of domains (visuo-motor learning, causal induction, problem solving, education, skill learning), the present study examines whether action-based learning is a necessary, or a sufficient, means of acquiring the relevant skills needed to perform a task typically described as requiring active learning. To achieve this, the present study compared the effects of action-based and observation-based learning on controlling a complex dynamic task environment. Both action- and observationbased learners either learnt by describing the changes in the environment in the form of a conditional statement, or not. The findings show that observational learners are sensitive to the instructional manipulations pursued during learning, in ways that are comparable to the active learning conditions. For both, advantages in performance, accuracy in knowledge of the task, and self-insight were found when learning was based on inducing rules from the task environment.

The article examines the question of how learning multiple tasks interacts with neural architectures and the flow of information through those architectures. It approaches the question by using the idealization of an artificial neural network where it is possible to ask more precise questions about the effects of modular versus nonmodular architectures as well as the effects of sequential versus simultaneous learning of tasks. A prior work has demonstrated a clear advantage of modular architectures when the two tasks must be learned at the same time from the start, but this advantage may disappear when one task is first learned to a criterion before the second task is undertaken. Indeed, in some cases of sequential learning, nonmodular networks achieve success levels comparable to those of modular networks. In particular, if a nonmodular network is to learn two tasks of different difficulty and the more difficult task is presented first and learned to a criterion, then the network will learn the second, easier one without permanent degradation of the first one. In contrast, if the easier task is learned first, a nonmodular task may perform significantly less well than a modular one. It seems that the reason for this difference has to do with the fact that the sequential presentation of the more difficult task first minimizes interference between the two tasks. More broadly, the studies summarized in this article seem to imply that no single learning architecture is optimal for all situations.

Marinov''s critique I argue, is vitiated by its failure to recognize the distinctive role of superposition within the distributed connectionist paradigm. The use of so-called subsymbolic distributed encodings alone is not, I agree, enough to justify treating distributed connectionism as a distinctive approach. It has always been clear that microfeatural decomposition is both possible and actual within the confines of recognizably classical approaches. When such approaches also involve statistically-driven learning algorithms — as in the case of ID3 — the fundamental differences become even harder to spot. To see them, it is necessary to consider not just the nature of an acquired input-output function but the nature of the representational scheme underlying it. Differences between such schemes make themselves best felt outside the domain of immediate problem solving. It is in the more extended contexts of performance DURING learning and cognitive change as a result of SUBSEQUENT training on new tasks (or simultaneous training on several tasks) that the effects of superpositional storage techniques come to the fore. I conclude that subsymbols, distribution and statistically driven learning alone are indeed not of the essence. But connectionism is not just about subsymbols and distribution. It is about the generation of whole subsymbol SYSTEMS in which multiple distributed representations are created and superposed.

This paper explores how an evolutionary process can produce systems that learn. A general framework for the evolution of learning is outlined, and is applied to the task of evolving mechanisms suitable for supervised learning in single-layer neural networks. Dynamic properties of a network’s information-processing capacity are encoded genetically, and these properties are subjected to selective pressure based on their success in producing adaptive behavior in diverse environments. As a result of selection and genetic recombination, various successful learning mechanisms evolve, including the well-known delta rule. The effect of environmental diversity on the evolution of learning is investigated, and the role of different kinds of emergent phenomena in genetic and connectionist systems is discussed.

It is widely appreciated that the difficulty of a particluar computation varies according to how the input data are presented. What is less understood is the effect of this computation/representation tradeoff within familiar learning paradigms. We argue that existing learning algoritms are often poorly equipped to solve problems involving a certain type of important and widespread regularity, which we call 'type-2' regularity. The solution in these cases is to trade achieved representation against computational search. We investigate several ways in which such a trade-off may be pursued including simple incremental learning, modular connectionism, and the developmental hypothesis of 'representational redescription'. In addition, the most distinctive features of human cognition- language and culture- may themselves be viewed as adaptions enabling this representation/computation trade-off to be pursued on an even grander scale.

We argue that existing learning algorithms are often poorly equipped to solve problems involving a certain type of important and widespread regularity that we call “type-2 regularity.” The solution in these cases is to trade achieved representation against computational search. We investigate several ways in which such a trade-off may be pursued including simple incremental learning, modular connectionism, and the developmental hypothesis of “representational redescription.”.