Sequential anomaly detection in the presence of noise and limited feedback

This paper describes a methodology for detecting anomalies from sequentially observed and potentially noisy data. The proposed approach consists of two main elements: 1) filtering, or assigning a belief or likelihood to each successive measurement based upon our ability to predict it from previous noisy observations and 2) hedging, or flagging potential anomalies by comparing the current belief against a time-varying and data-adaptive threshold. The threshold is adjusted based on the available feedback from an end user. Our algorithms, which combine universal prediction with recent work on online convex programming, do not require computing posterior distributions given all current observations and involve simple primal-dual parameter updates. At the heart of the proposed approach lie exponential-family models which can be used in a wide variety of contexts and applications, and which yield methods that achieve sublinear per-round regret against both static and slowly varying product distributions with marginals drawn from the same exponential family. Moreover, the regret against static distributions coincides with the minimax value of the corresponding online strongly convex game. We also prove bounds on the number of mistakes made during the hedging step relative to the best offline choice of the threshold with access to all estimated beliefs and feedback signals. We validate the theory on synthetic data drawn from a time-varying distribution over binary vectors of high dimensionality, as well as on the Enron email dataset. © 1963-2012 IEEE.

Bayesian variable selection in structured high-dimensional covariate spaces with applications in genomics

We consider the problem of variable selection in regression modeling in high-dimensional spaces where there is known structure among the covariates. This is an unconventional variable selection problem for two reasons: (1) The dimension of the covariate space is comparable, and often much larger, than the number of subjects in the study, and (2) the covariate space is highly structured, and in some cases it is desirable to incorporate this structural information in to the model building process. We approach this problem through the Bayesian variable selection framework, where we assume that the covariates lie on an undirected graph and formulate an Ising prior on the model space for incorporating structural information. Certain computational and statistical problems arise that are unique to such high-dimensional, structured settings, the most interesting being the phenomenon of phase transitions. We propose theoretical and computational schemes to mitigate these problems. We illustrate our methods on two different graph structures: the linear chain and the regular graph of degree k. Finally, we use our methods to study a specific application in genomics: the modeling of transcription factor binding sites in DNA sequences. © 2010 American Statistical Association.

Relative growth of the limbs and trunk in sifakas: heterochronic, ecological, and functional considerations.

Limb, trunk, and body weight measurements were obtained for growth series of Milne-Edwards's diademed sifaka, Propithecus diadema edwardsi, and the golden-crowned sifaka, Propithecus tattersalli. Similar measures were obtained also for primarily adults of two subspecies of the western sifaka: Propithecus verreauxi coquereli, Coquerel's sifaka, and Propithecus verreauxi verreauxi, Verreaux's sifaka. Ontogenetic series for the larger-bodied P. d. edwardsi and the smaller-bodied P. tattersalli were compared to evaluate whether species-level differences in body proportions result from the differential extension of common patterns of relative growth. In bivariate plots, both subspecies of P. verreauxi were included to examine whether these taxa also lie along a growth trajectory common to all sifakas. Analyses of the data indicate that postcranial proportions for sifakas are ontogenetically scaled, much as demonstrated previously with cranial dimensions for all three species (Ravosa, 1992). As such, P. d. edwardsi apparently develops larger overall size primarily by growing at a faster rate, but not for a longer duration of time, than P. tattersalli and P. verreauxi; this is similar to results based on cranial data. A consideration of Malagasy lemur ecology suggests that regional differences in forage quality and resource availability have strongly influenced the evolutionary development of body-size variation in sifakas. On one hand, the rainforest environment of P. d. edwardsi imposes greater selective pressures for larger body size than the dry-forest environment of P. tattersalli and P. v. coquereli, or the semi-arid climate of P. v. verreauxi. On the other hand, as progressively smaller-bodied adult sifakas are located in the east, west, and northwest, this apparently supports suggestions that adult body size is set by dry-season constraints on food quality and distribution (i.e., smaller taxa are located in more seasonal habitats such as the west and northeast). Moreover, the fact that body-size differentiation occurs primarily via differences in growth rate is also due apparently to differences in resource seasonality (and juvenile mortality risk in turn) between the eastern rainforest and the more temperate northeast and west. Most scaling coefficients for both arm and leg growth range from slight negative allometry to slight positive allometry. Given the low intermembral index for sifakas, which is also an adaptation for propulsive hindlimb-dominated jumping, this suggests that differences in adult limb proportions are largely set prenatally rather than being achieved via higher rates of postnatal hindlimb growth.(ABSTRACT TRUNCATED AT 400 WORDS)

The abelianization of the Johnson kernel

We prove that the first complex homology of the Johnson subgroup of the Torelli group Tg is a non-trivial, unipotent Tg-module for all g ≥ 4 and give an explicit presentation of it as a Sym H 1(Tg,C)-module when g ≥ 6. We do this by proving that, for a finitely generated group G satisfying an assumption close to formality, the triviality of the restricted characteristic variety implies that the first homology of its Johnson kernel is a nilpotent module over the corresponding Laurent polynomial ring, isomorphic to the infinitesimal Alexander invariant of the associated graded Lie algebra of G. In this setup, we also obtain a precise nilpotence test. © European Mathematical Society 2014.

A theory of hypoellipticity and unique ergodicity for semilinear stochastic PDEs

We present a theory of hypoellipticity and unique ergodicity for semilinear parabolic stochastic PDEs with "polynomial" nonlinearities and additive noise, considered as abstract evolution equations in some Hilbert space. It is shown that if Hörmander's bracket condition holds at every point of this Hilbert space, then a lower bound on the Malliavin covariance operatorμt can be obtained. Informally, this bound can be read as "Fix any finite-dimensional projection on a subspace of sufficiently regular functions. Then the eigenfunctions of μt with small eigenvalues have only a very small component in the image of Π." We also show how to use a priori bounds on the solutions to the equation to obtain good control on the dependency of the bounds on the Malliavin matrix on the initial condition. These bounds are sufficient in many cases to obtain the asymptotic strong Feller property introduced in [HM06]. One of the main novel technical tools is an almost sure bound from below on the size of "Wiener polynomials," where the coefficients are possibly non-adapted stochastic processes satisfying a Lips chitz condition. By exploiting the polynomial structure of the equations, this result can be used to replace Norris' lemma, which is unavailable in the present context. We conclude by showing that the two-dimensional stochastic Navier-Stokes equations and a large class of reaction-diffusion equations fit the framework of our theory.