
	Where results make sense

Topic: Supervised learning

In the News (Wed 8 Oct 08)

Breast Cancer

Golden Circle

Keating Five

Stephen Harper

Eli Lilly

Nobel Prize

Eli Manning

Josh Beckett

Sage Rosenfels

South Ossetia

	Reinforcement Learning
		Thus while much of the focus of machine learning has been on supervised learning, if we are to understand learning in nature, we need to study unsupervised and reinforcement learning.
		That is, before learning, the agent may not know what will happen when it takes a particular action in a particular state, but the only relevant information for deciding what action to take is the current state, which the agent does have access to.
		Early in learning, it is better to explore because the knowledge the agent has gained so far is not very reliable and because a number of options may still need to be tried.
	www.cs.indiana.edu /~gasser/Salsa/rl.html (2197 words)

	Supervised learning - Wikipédia
		Supervised learning is a machine learning technique for creating a function from training data.
		Parameters of the learning algorithm may be adjusted by optimizing performance on a subset (called a validation set) of the training set, or via cross-validation.
		After parameter adjustment and learning, the performance of the algorithm may be measured on a test set that is separate from the training set.
	su.wikipedia.org /wiki/Supervised_learning (370 words)

	Learning - Wikipedia, the free encyclopedia
		Learning is the process of acquiring knowledge, skills, attitudes, or values, through study, experience, or teaching, that causes a change of behavior that is persistent, measurable, and specified or allows an individual to formulate a new mental construct or revise a prior mental construct (conceptual knowledge such as attitudes or values).
		Learning "how to learn" is a skill, which can be taught to others, by example.
		Learning by teaching, is a method of teaching which allows teachers to share the new lesson contents and let little groups prepare their part in order to teach this contents to the rest of the class.
	en.wikipedia.org /wiki/Learning (827 words)

	Supervised learning - Wikipedia, the free encyclopedia
		In some cases, however, the map is implemented as a set of local models (such as in case-based reasoning or the nearest neighbor algorithm).
		The goal of supervised learning of a global model is to find a function g, given a set of points of the form (x, g(x)).
		Statistical learning theory investigates under what conditions empirical risk minimization is admissible and how good the approximations can be expected to be.
	en.wikipedia.org /wiki/Supervised_learning (681 words)

	Encyclopedia: Supervised learning
		Statistical classification is a type of supervised learning problem in which labeled training data is used to create a function that will correctly predict the label of future data.
		Unsupervised learning is a method of machine learning where a model is fit to observations.
		Probably approximately correct learning (PAC learning) is a framework of learning that was proposed by Leslie Valiant in his paper A theory of the learnable.
	www.nationmaster.com /encyclopedia/Supervised-learning (1794 words)

	Machine learning - Wikipedia, the free encyclopedia
		Machine learning algorithms are organized into a taxonomy, based on the desired outcome of the algorithm.
		supervised learning --- where the algorithm generates a function that maps inputs to desired outputs.
		One standard formulation of the supervised learning task is the classification problem: the learner is required to learn (to approximate the behavior of) a function which maps a vector
	en.wikipedia.org /wiki/Machine_learning (807 words)

	Learning
		Used in supervised learning, a training set is a set of problem instances (described as a set of properties and their values), together with a classification of the instance.
		Learning in which the data structure is a set of nodes connected by weighted links, each node passing a 0 or 1 to other links depending on whether a function of its inputs reaches its activation level.
		The reason reinforcement learning is harder than supervised learning is that the agent is never told what the right action is, only whether it is doing well or poorly, and in some cases (such as chess) it may only receive feedback after a long string of actions.
	www.cs.dartmouth.edu /~brd/Teaching/AI/Lectures/Summaries/learning.html (9712 words)

	1.1 Reinforcement Learning
		Reinforcement learning is learning what to do--how to map situations to actions--so as to maximize a numerical reward signal.
		A full specification of the reinforcement learning problem in terms of optimal control of Markov decision processes must wait until Chapter 3, but the basic idea is simply to capture the most important aspects of the real problem facing a learning agent interacting with its environment to achieve a goal.
		For learning research to make progress, important subproblems have to be isolated and studied, but they should be subproblems that play clear roles in complete, interactive, goal-seeking agents, even if all the details of the complete agent cannot yet be filled in.
	www.cs.ualberta.ca /~sutton/book/ebook/node7.html (956 words)

	Supervised Learning
		"Supervised learning is a machine learning technique whereby the algorithm is first presented with training data which consists of examples which include both the inputs and the desired outputs; thus enabling it to learn a function.
		The task of the supervised learner is to predict the value of the function for any valid input object after having seen a number of training examples (i.e.
		"In supervised learning, a teacher provides a category label or cost for each pattern in a training set, and seeks to reduce the sum of the costs for these patterns.
	www.machinelearning.net /supervised-learning (308 words)

	Categorical Perception and the Evolution of Supervised Learning in Neural Nets (Site not responding. Last check: 2007-10-17)
		CP effects have also been reported to occur purely as a result of learning in experiments with artificial continua; similar "warping" effects might be expected to arise from learning complex multidimensional categories, as in learning to sort baby chicks as male and female, or histological slides as cancerous or noncancerous.
		They acquire the capacity to sort their inputs into the categories imposed by supervised learning through altering the pairwise distances between them (where distance is the degree to which a pair of inputs is discriminable by the net) until there is sufficient within-category compression and between-category separation to accomplish reliable categorization.
		Supervision in the form of feedback from the consequences of mis categorization may be our best guide as to how to carve up objects, events and states of affairs.
	cogsci.soton.ac.uk /harnad/Papers/Harnad/harnad91.cpnets.html (3423 words)

	Supervised Learning: Are We Done Yet? (Site not responding. Last check: 2007-10-17)
		In supervised learning, the goal is to learn a function that maps from the input variables (e.g., patient descriptors, or the words in a document) to an output classification (this document is a homepage) or a predicted output value (this patient has risk 0.82).
		This function is learned from a training set that includes the input variables and outputs to be predicted for each case.
		Supervised learning has been around for 40+ years (longer if we count regression), and we now have more than two dozen different supervised learning algorithms.
	www.cs.cornell.edu /~ves/Abstracts/RichCaruana.htm (171 words)

	Supervised learning -- Facts, Info, and Encyclopedia article (Site not responding. Last check: 2007-10-17)
		Supervised learning is a (Click link for more info and facts about machine learning) machine learning technique for creating a function from training data.
		In some cases, however, the map is implemented as a set of local models (such as in (Click link for more info and facts about case-based reasoning) case-based reasoning or the (Click link for more info and facts about nearest neighbor algorithm) nearest neighbor algorithm).
		Parameters of the learning algorithm may be adjusted by optimizing performance on a subset (called a validation set) of the training set, or via (Click link for more info and facts about cross-validation) cross-validation.
	www.absoluteastronomy.com /encyclopedia/s/su/supervised_learning.htm (1143 words)

	Learning Rules
		Supervised learning means that target values for the output are presented to the network, in order that the network can update its weights such that the euclidean distance of the output vector and the target vector is minimised.
		Supervised learning attempts to match the output of the network to values that have already been defined.
		By further supervised learning using LVQ, a two stage learning adaptively is used to solve a classification problem.
	openai.sourceforge.net /docs/nn_algorithms/networksarticle/rules.html (2087 words)

	Text Learning: Beyond Supervision (Site not responding. Last check: 2007-10-17)
		This research is especially relevant to text learning because the inherent complexity of natural language usually requires models with many parameters: estimating those parameters requires a lot of data, and obtaining labeled data is difficult and expensive.
		By combining unsupervised learning with supervised learning, the need for labeled training data can often be greatly reduced, allowing for the development of more powerful models and methods.
		The goal of this workshop is to bring together researchers from the machine learning, information retrieval, computational linguistics and language modeling to talk about their different perspectives on this intersection and to share their latest ideas.
	www.cs.cmu.edu /~mccallum/textbeyond (329 words)

	Manifold Learning Resource Page
		Costa and A. Hero, Geodesic entropic graphs for dimension and entropy estimation in manifold learning.
		Spectral Clustering and Kernel PCA are Learning Eigenfunctions.
		Learning with Cascade for Classification of Non-Convex Manifolds.
	www.cse.msu.edu /~lawhiu/manifold (2060 words)

	IRI WEB SITE - PROJECT:
		The system should be able to learn its surroundings in a supervised form, as well as the more relevant objects that allow for the completion of a particular task.
		Once a map of the environment has been created, and the relevant objects have been learned, the robot should be able to navigate through the environment identifying the learned objects; even if there have been significant changes in the environment from the learning phase to the recognition phase.
		This project is the continuation of a previous CICYT project assigned to the main investigator on 1998, in which we developed robust learning techniques for a mobile robot called MARCO using active vision.In this project we will develop robust techniques for the extraction and learning of invariant image features based on color, texture, and geometry.
	www-iri.upc.es /english/mostraproject.php?proj=40 (495 words)

	Journal of College Student Development: Learning Through Supervised Practice in Student Affairs
		Supervised practice and theory to practice has long been a focus in the professional literature (Argyris & Schön, 1974), and several authors of this already text have made significant contributions to this literature base (Winston, Creamer, & Miller, 2001).
		Given the remarkably broad and deep scholarly and professional experience of the authors, the text not only provides current state of the supervised practice art, but in many ways is the culmination.
		In sum, Learning Through Supervised Practice can serve not only as a primer to the new professional, but a reference book for seasoned professionals looking for a refresher.
	www.findarticles.com /p/articles/mi_qa3752/is_200407/ai_n9457045 (870 words)

	Supervised Learning and Classification (Site not responding. Last check: 2007-10-17)
		Supervised Classification is a special case of the above where U is a discrete attribute often called "the class".
		Overfitting: When a too complex model is learned - often revealed when errors on the test set are much worse than those on the training set.
		Supervised classification is a special case of supervised learning where the output-space, U, is a single discrete attribute often called "the class".
	www.csse.monash.edu.au /~lloyd/Archive/2004-07-Supervised (195 words)

	What is Semi-Supervised Learning? (Site not responding. Last check: 2007-10-17)
		As the name suggests semi-supervised learning is the middle road between supervised and unsupervised learning.
		In supervised learning, programs are trained under complete supervision, meaning a neural network is forced to learn every training pattern correctly.
		In unsupervised learning, training is not supervised and patterns are clustered based on their similarity.
	pegasus.cc.ucf.edu /~mbharadw/thesis/semi_supervised_learning.htm (408 words)

	Semi-Supervised Learning Survey (Site not responding. Last check: 2007-10-17)
		Unlike the thesis, the survey is updated from time to time to incorporate latest developments in the field of semi-supervised learning.
		It can be used as a short tutorial on semi-supervised learning methods.
		The goal of semi-supervised learning is to train better classifiers from both labeled and unlabeled data.
	www.cs.wisc.edu /~jerryzhu/research/ssl/semireview.html (158 words)

	Q530: Reinforcement Learning
		A third alternative, much closer to supervised than unsupervised learning, is reinforcement learning: the learner receives feedback (sometimes) about the appropriateness of its response.
		We will divide the agent's policy into two components, how good the agent thinks an action is for a given state and how the agent uses this knowledge to choose an action for a given state.
		That is, learn to take actions that lead you to that state.
	www.indiana.edu /~gasser/Q530/Notes/rl.html (2355 words)

	Supervised Learning and Data Mining (Tom Dietterich)
		Supervised learning from examples involves learning a function y = f(x) from training examples of the form (x, f(x)).
		The learning algorithms used to train the "base classifier" (i.e., the classifier applied to the individual pieces) are not designed for the divide-and-conquer setting, and the "merge" methods are typically not trained at all.
		A goal of our research is to develop learning algorithms for the "base classifier" and "merge" steps that can be applied to a wide variety of divide-and-conquer learning algorithms.
	web.engr.oregonstate.edu /~tgd/projects/supervised.html (1082 words)

	[No title]
		An important distinction in machine learning, and also applicable to data mining, is that between supervised and unsupervised learning algorithms.
		The term "supervised" learning is usually applied to cases in which a particular classification is already observed and recorded in a training sample, and you want to build a model to predict those classifications (in a new testing sample).
		These methods are called supervised learning algorithms because the learning (fitting of models) is "guided" or "supervised" by the observed classifications recorded in the data file.
	www.statsoftinc.com /textbook/gloss.html (8802 words)

	Temporal Difference Learning and TD-Gammon
		The reinforcement learning paradigm has held great intuitive appeal and has attracted considerable interest for many years because of the notion of the learner being able to learn on its own, without the aid of an intelligent "teacher," from its own experience at attempting to perform a task.
		Another problem with many of the traditional approaches to reinforcement learning is that they have been limited to learning either lookup tables or linear evaluation functions, neither of which seem adequate for handling many classes of real-world problems.
		Finally, non-deterministic games have the advantage that the target function one is trying to learn, the true expected outcome of a position given perfect play on both sides, is a real-valued function with a great deal of smoothness and continuity, that is, small changes in the position produce small changes in the probability of winning.
	www.research.ibm.com /massive/tdl.html (7500 words)

	BioMed Central \| Full text \| Phylogenetic tree information aids supervised learning for predicting protein-protein ...
		Both the unsupervised and supervised learning paradigms benefit from the explicit inclusion of these intermediate distance matrices, and particularly so in the latter case, which offers a better balance between sensitivity and specificity in the prediction of protein-protein interactions.
		Second, in contrast to the unsupervised learning scheme of the mirror tree method, we adopt a supervised learning paradigm, specifically the support vector machines, [20-22] to further tap into the prior knowledge about interacting and non interacting protein pairs.
		To further examine the performance, in Figure 4, ROC curves are shown for those ROC scores reported in Table 3 for the unsupervised learning based on Pearson correlation coefficient and for the supervised learning with a Gaussian kernel SVM.
	www.biomedcentral.com /1471-2105/8/6 (5759 words)

	Supervised learning (Site not responding. Last check: 2007-10-17)
		These two types of data (the representation of the objects and their responses in the system) form pairs which for the present purpose are called inputs and targets.
		Thus the basic idea in these supervised learning techniques is that there are minimally 4 data sets to be studied, as follows.
		As in all other data analysis techniques, these supervised learning methods are not immune from sensitivity to badly chosen initial data (Zupan & Gasteiger, 1993).
	qbab.aber.ac.uk /roy/supervis/suprlrn.htm (676 words)

Try your search on: Qwika (all wikis)