1,720,998 research outputs found
Transfer of samples in batch reinforcement learning
No full textThe main objective of transfer in reinforcement learning is to reduce the complexity of learning the solution of a target task by effectively reusing the knowledge retained from solving a set of source tasks. In this paper, we introduce a novel algorithm that transfers samples (i.e., tuples ) from source to target tasks. Under the assumption that tasks have similar transition models and reward functions, we propose a method to select samples from the source tasks that are mostly similar to the target task, and, then, to use them as input for batch reinforcement-learning algorithms. As a result, the number of samples an agent needs to collect from the target task to learn its solution is reduced. We empirically show that, following the proposed approach, the transfer of samples is effective in reducing the learning complexity, even when some source tasks are significantly different from the target task
Hidden Markov Models
No full textF53.36> a 0 which represents the null output -- in other words if the model generates the symbol a 0 it simply does not output anything. Each state s j has an output distribution defined by the vector A j such that the probability of emitting symbol a when we are in state s j is given by A j (a). Of course we require that: X a2A A j (a) = 1 8s j 2 S There is also a matrix T of transistion probabilities defined so that T jk is the probability of moving into state s k if the model is currently in state s j . Note that rows of T must sum to unity (although columns may not) and also that the diagonal elements of<F53
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Constrained hidden Markov models
No full textBy thinking of each state in a hidden Markov model as corresponding to some spatial region of a fictitious topology space it is possible to naturally define neighbouring states as those which are connected in that space. The transition matrix can then be constrained to allow transitions only between neighbours; this means that all valid state sequences correspond to connected paths in the topology space. I show how such constrained HMMs can learn to discover underlying structure in complex sequences of high dimensional data, and apply them to the problem of recovering mouth movements from acoustics in continuous speech. 1 Latent variable models for structured sequence data Structured time-series are generated by systems whose underlying state variables change in a continuous way but whose state to output mappings are highly nonlinear, many to one and not smooth. Probabilistic unsupervised learning for such sequences requires models with two essential features: latent (hidden) variables and topology in those variables. Hidden Markov models (HMMs) can be thought of as dynamic generalizations of discrete stat
Speech Processing Background
No full textIntroduction This note provides an extremely brief and necessarily incomplete introduction to speech processing by machines for those unfamiliar with the basics of the eld. It is clearly beyond the scope of such a tutorial to give a comprehensive survey of computer speech processing methods. Below I provide a very general overview of the current paradigms used in speech processing. I do not, however, provide details of implementing a recognition system; although unfortunately most of the work in getting a system to actually function properly is in the details. I also give examples of state of the art performance for recognition, synthesis, speaker identi cation and compression systems. Three standard textbooks, one old but classic (by Rabiner and Schafer [42]) and two newer (by Rabiner and Juang [41] and Deller , Proakis and Hansen [14]), provide very comprehensive introductions to this material. The collection edited by Waibel [48] provides an excellent source of important early p
1 What Are HMMs? Hidden Markov Models
No full textHidden Markov Models (HMMs) are a class of models for mimicing the probability density of a sequence of observed symbols. They are essentially stochastic nite state machines which output asymbol eachtime they depart from a state. By specifying the state transistion probabilities between states and the symbol generation model for each state, we can attempt to capture the underlying structureinalargesetofsymbol strings. In general, the operational paradigm is as follows: select a starting state according tosome xed probability distribution. At eachtime step, generate anoutput symbol byinvoking the generative model of the current state, and then transition to anew state according toastatic transistion probability matrix. Let us de ne some notation for future convenience. Assume there are N possible states sn in our model, and denote the set of all possible states by S where S = fs1;s2;:::;sNg. To this set, we will add a special state s0 in which the model always starts asanimplementational convenience { the use of this state will be explained later. We may choose to useoneofthe states (usually sN) asanend state {inthis case, whenever the model reaches this state, it stops. Let there be P possible output symbols ap which comprise the output alphabet A = fa1;a2;:::;aP g. To this set, we addthe special symbol a0 which represents the null output { in other words if the model generates the symbol a0 it simply does not output anything. Each state sj has an output distribution de ned by the vector Aj such that the probability ofemitting symbol a when we are in state sj is given by Aj(a). Of course we require that: X a2
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
