arXiv subjects

Timothy H. Keitt

Publications and source records attributed to Timothy H. Keitt.

4 recordsLinked to original sources

arXiv2026

Adaptive Evidence Weighting for Audio-Spatiotemporal Fusion

Many machine learning systems have access to multiple sources of evidence for the same prediction target, yet these sources often differ in reliability and informativeness across inputs. In bioacoustic classification, species identity may be inferred both from the acoustic signal and from spatiotemporal context such as location and season; while Bayesian inference motivates multiplicative evidence combination, in practice we typically only have access to discriminative predictors rather than calibrated generative models. We introduce \textbf{F}usion under \textbf{IN}dependent \textbf{C}onditional \textbf{H}ypotheses (\textbf{FINCH}), an adaptive log-linear evidence fusion framework that integrates a pre-trained audio classifier with a structured spatiotemporal predictor. FINCH learns a per-sample gating function that estimates the reliability of contextual information from uncertainty and informativeness statistics. The resulting fusion family \emph{contains} the audio-only classifier as a special case and explicitly bounds the influence of contextual evidence, yielding a risk-contained hypothesis class with an interpretable audio-only fallback. Across benchmarks, FINCH consistently outperforms fixed-weight fusion and audio-only baselines, improving robustness and error trade-offs even when contextual information is weak in isolation. We achieve state-of-the-art performance on CBI and competitive or improved performance on several subsets of BirdSet using a lightweight, interpretable, evidence-based approach. Code is available: \texttt{\href{https://anonymous.4open.science/r/birdnoise-85CD/README.md}{anonymous-repository}}

cs.SD↗

arXiv2025

Spatial Hyperspheric Models for Compositional Data

Compositional observations are an increasingly prevalent data source in spatial statistics. Analysis of such data is typically done on log-ratio transformations or via Dirichlet regression. However, these approaches often make unnecessarily strong assumptions (e.g., strictly positive components, exclusively negative correlations). An alternative approach uses square-root transformed compositions and directional distributions. Such distributions naturally allow for zero-valued components and positive correlations, yet they may include support outside the non-negative orthant and are not generative for compositional data. To overcome this challenge, we truncate the elliptically symmetric angular Gaussian (ESAG) distribution to the non-negative orthant. Additionally, we propose a spatial hyperspheric regression model that contains fixed and random multivariate spatial effects. The proposed model also contains a term that can be used to propagate uncertainty that may arise from precursory stochastic models (i.e., machine learning classification). We used our model in a simulation study and for a spatial analysis of classified bioacoustic signals of the Dryobates pubescens (downy woodpecker).

stat.ME↗

arXiv2010

A sampling theory for asymmetric communities

We introduce the first analytical model of asymmetric community dynamics to yield Hubbell's neutral theory in the limit of functional equivalence among all species. Our focus centers on an asymmetric extension of Hubbell's local community dynamics, while an analogous extension of Hubbell's metacommunity dynamics is deferred to an appendix. We find that mass-effects may facilitate coexistence in asymmetric local communities and generate unimodal species abundance distributions indistinguishable from those of symmetric communities. Multiple modes, however, only arise from asymmetric processes and provide a strong indication of non-neutral dynamics. Although the exact stationary distributions of fully asymmetric communities must be calculated numerically, we derive approximate sampling distributions for the general case and for nearly neutral communities where symmetry is broken by a single species distinct from all others in ecological fitness and dispersal ability. In the latter case, our approximate distributions are fully normalized, and novel asymptotic expansions of the required hypergeometric functions are provided to make evaluations tractable for large communities. Employing these results in a Bayesian analysis may provide a novel statistical test to assess the consistency of species abundance data with the neutral hypothesis.

q-bio.PE↗

arXiv2001

Scaling in the growth of geographically subdivided populations: invariant patterns from a continent-wide biological survey

We consider statistical patterns of variation in growth rates for over 400 species of breeding birds across North America surveyed from 1966 to 1998. We report two results. First, the standard deviation of population growth rates decays as a power-law function of total population size with an exponent $β= 0.36\pm0.02$. Second, the number of subpopulations, measured as the number of survey locations with non-zero counts, scales to the 3/4-power of total number of birds counted in a given species. We show how these patterns may be related, and discuss a simple stochastic growth model for a geographically subdivided population that formalizes the relationship. We also examine reasons that may explain why some species deviate from these scaling-laws.

cond-mat.stat-mech↗