Results 1 - 10 of 1,185

Exploiting Generative Models in Discriminative Classifiers

by Tommi Jaakkola, David Haussler - In Advances in Neural Information Processing Systems 11 , 1998
"... Generative probability models such as hidden Markov models provide a principled way of treating missing information and dealing with variable length sequences. On the other hand, discriminative methods such as support vector machines enable us to construct flexible decision boundaries and often resu ..."
Abstract - Cited by 551 (9 self) - Add to MetaCart
result in classification performance superior to that of the model based approaches. An ideal classifier should combine these two complementary approaches. In this paper, we develop a natural way of achieving this combination by deriving kernel functions for use in discriminative methods such as support

Graphs over Time: Densification Laws, Shrinking Diameters and Possible Explanations

by Jure Leskovec, Jon Kleinberg, Christos Faloutsos , 2005
"... How do real graphs evolve over time? What are “normal” growth patterns in social, technological, and information networks? Many studies have discovered patterns in static graphs, identifying properties in a single snapshot of a large network, or in a very small number of snapshots; these include hea ..."
Abstract - Cited by 541 (48 self) - Add to MetaCart
increase slowly as a function of the number of nodes (like O(log n) orO(log(log n)). Existing graph generation models do not exhibit these types of behavior, even at a qualitative level. We provide a new graph generator, based on a “forest fire” spreading process, that has a simple, intuitive justification

Likelihood-based belief function: justification and some extensions to low-quality data

by Thierry Denœux
"... Given a parametric statistical model, evidential methods of statistical in-ference aim at constructing a belief function on the parameter space from observations. The two main approaches are Dempster’s method, which re-gards the observed variable as a function of the parameter and an auxiliary varia ..."
Abstract - Add to MetaCart
Given a parametric statistical model, evidential methods of statistical in-ference aim at constructing a belief function on the parameter space from observations. The two main approaches are Dempster’s method, which re-gards the observed variable as a function of the parameter and an auxiliary

Justification

by Caqca Tracking Prorizm, Y Nelson, Aixt/gst/zns/x- Electesaprs E, Best Available Copy, A Fuzzy, Logic Optna, L Comirlm, Law Solutnon, Y Nelson, B. S. Bydistribution I, Iordis Spechd , 1993
"... The purpose of this thesis wan to develop a set of control laws, impleme~nted in a simulation, to allow the Cruise Missile Control Aircraft (CXMCA) to radar track cruise missiles and achieve 100 % radar coverage during test flights. The CXNCA was forced to perform a series of intercepts on set point ..."
Abstract - Add to MetaCart
points recomputed every ton seconds. Fuszy logic formed a vital function in determining wbore the set point would be, and used current and future missile posit4 on data. j This thosis development and completion was only possible through the tireless help and motivation provi4e by my advisor, Col Don

Graph evolution: Densification and shrinking diameters

by Jure Leskovec, Jon Kleinberg, Christos Faloutsos - ACM TKDD , 2007
"... How do real graphs evolve over time? What are “normal” growth patterns in social, technological, and information networks? Many studies have discovered patterns in static graphs, identifying properties in a single snapshot of a large network, or in a very small number of snapshots; these include hea ..."
Abstract - Cited by 267 (16 self) - Add to MetaCart
increase slowly as a function of the number of nodes (like O(log n) or O(log(log n)). Existing graph generation models do not exhibit these types of behavior, even at a qualitative level. We provide a new graph generator, based on a “forest fire” spreading process, that has a simple, intuitive

The Security of the Cipher Block Chaining Message Authentication Code

by Mihir Bellare, Joe Kilian , Phillip Rogaway , 2000
"... Let F be some block cipher (eg., DES) with block length l. The Cipher Block Chaining Message Authentication Code (CBC MAC) specifies that an m-block message x: Xl...xm be authenticated among parties who share a secret key a for the block cipher by tagging x with a prefix of ym, where Y0: 01 and Y ..."
Abstract - Cited by 240 (41 self) - Add to MetaCart
and Yi: Fa(miYi-1) for i: 1,2,...,m. This method is a pervasively used international and U.S. standard. We provide its first formal justification, showing the following general lemma: cipher block chaining a pseudorandom function yields a pseudorandom function. Underlying our results is a technical

1 The Justification of Financial Futures Exchanges

by David Campbell, Sol Picciotto
"... The invention of exchange-traded financial futures by commodity exchanges has been justified by Chicagoan financial economics and law and economics on the basis of the unrealistic assumption that trading is costless. Acceptance of this quite circular argument has led to the delegation of extensive p ..."
Abstract - Add to MetaCart
‘freedom’, there should be stronger public consideration of the optimal public-private mix of regulatory governance structures. This paper proposes criteria for the evaluation of the functional justification of exchange-traded futures as a guide for public policy aimed at reducing the volatility

1.5 Summary and justification

by Elselijn Kingma, Dispositional Function
"... Christopher Boorse’s Bio Statistical Theory (BST) defines health as the absence of disease, and disease as the adverse departure from normal species functioning. This paper presents a two-pronged problem for this account. First I demonstrate that, in order to accurately account for dynamic physiolog ..."
Abstract - Add to MetaCart
Christopher Boorse’s Bio Statistical Theory (BST) defines health as the absence of disease, and disease as the adverse departure from normal species functioning. This paper presents a two-pronged problem for this account. First I demonstrate that, in order to accurately account for dynamic

Semi-Supervised Learning on Riemannian Manifolds

by Mikhail Belkin , Partha Niyogi , 2004
"... We consider the general problem of utilizing both labeled and unlabeled data to improve classification accuracy. Under the assumption that the data lie on a submanifold in a high dimensional space, we develop an algorithmic framework to classify a partially labeled data set in a principled manner. ..."
Abstract - Cited by 193 (7 self) - Add to MetaCart
. The central idea of our approach is that classification functions are naturally defined only on the submanifold in question rather than the total ambient space. Using the Laplace-Beltrami operator one produces a basis (the Laplacian Eigenmaps) for a Hilbert space of square integrable functions

ON JUSTIFICATION OF GIBBS DISTRIBUTION 1

by unknown authors , 2001
"... The paper develop a new approach to the justification of Gibbs canonical distribution for Hamiltonian systems with finite number of degrees of freedom. It uses the condition of nonintegrability of the ensemble of weak interacting Hamiltonian systems. 1. Gibbs distribution. We consider the probabilit ..."
Abstract - Add to MetaCart
The paper develop a new approach to the justification of Gibbs canonical distribution for Hamiltonian systems with finite number of degrees of freedom. It uses the condition of nonintegrability of the ensemble of weak interacting Hamiltonian systems. 1. Gibbs distribution. We consider
Results 1 - 10 of 1,185