arXiv subjects

Eric Taylor

Publications and source records attributed to Eric Taylor.

3 recordsLinked to original sources

arXiv2026

Retrieval-augmented generation vs. deterministic tax computation in multi-agent financial advisory: A 2x2 factorial experiment

Tax-loss harvesting demonstrates consistent benefits to long-term portfolio growth; yet implementing it efficiently often involves complex considerations that are specific to the holdings within that portfolio and the individual who owns it. We introduce a custom capital gains calculation engine and a RAG-retrieved vector store of market advisory reports to provide context for a multi-agent trade recommendation system. We investigate the effects of each context provider on the quality of recommendations, measured by relative capital gains incurred during portfolio liquidation. A 2x2 repeated-measures ANOVA revealed a significant main effect of the tax optimization engine ($F(1,29) = 9.17$, $p = .005$, $\eta^2_p = .240$): enabling the engine reduced tax savings by approximately 55 percentage points relative to the no-engine conditions. The RAG main effect was not significant ($p = .841$), nor was the interaction ($p = .553$). The RAG-only condition achieved the highest descriptive mean tax savings (47.7%), and the baseline condition performed second-best (30.6%), suggesting that the pre-trained language model's internalized financial knowledge may be sufficient for competent tax-loss harvesting recommendations without explicit tooling. These results indicate that augmenting LLM agents with domain-specific computation engines does not guarantee improved performance and may introduce conflicting optimization signals.

cs.AI↗

arXiv2024

Parameter Estimation and Identifiability in Kinetic Flux Profiling Models of Metabolism

Metabolic fluxes are the rates of life-sustaining chemical reactions within a cell and metabolites are the components. Determining the changes in these fluxes is crucial to understanding diseases with metabolic causes and consequences. Kinetic flux profiling (KFP) is a method for estimating flux that utilizes data from isotope tracing experiments. In these experiments, the isotope-labeled nutrient is metabolized through a pathway and integrated into the downstream metabolite pools. Measurements of proportion labeled for each metabolite in the pathway are taken at multiple time points and used to fit an ordinary differential equations model with fluxes as parameters. We begin by generalizing the process of converting diagrams of metabolic pathways into mathematical models composed of differential equations and algebraic constraints. The scaled differential equations for proportions of unlabeled metabolite contain parameters related to the metabolic fluxes in the pathway. We investigate flux parameter identifiability given data collected only at the steady state of the differential equation. Next, we give criteria for valid parameter estimations in the case of a large separation of timescales with fast-slow analysis. Bayesian parameter estimation on simulated data from KFP experiments containing both irreversible and reversible reactions illustrates the accuracy and reliability of flux estimations. These analyses provide constraints that serve as guidelines for the design of KFP experiments to estimate metabolic fluxes.

math.DS↗

arXiv2020

Neuron-based explanations of neural networks sacrifice completeness and interpretability

High quality explanations of neural networks (NNs) should exhibit two key properties. Completeness ensures that they accurately reflect a network's function and interpretability makes them understandable to humans. Many existing methods provide explanations of individual neurons within a network. In this work we provide evidence that for AlexNet pretrained on ImageNet, neuron-based explanation methods sacrifice both completeness and interpretability compared to activation principal components. Neurons are a poor basis for AlexNet embeddings because they don't account for the distributed nature of these representations. By examining two quantitative measures of completeness and conducting a user study to measure interpretability, we show the most important principal components provide more complete and interpretable explanations than the most important neurons. Much of the activation variance may be explained by examining relatively few high-variance PCs, as opposed to studying every neuron. These principal components also strongly affect network function, and are significantly more interpretable than neurons. Our findings suggest that explanation methods for networks like AlexNet should avoid using neurons as a basis for embeddings and instead choose a basis, such as principal components, which accounts for the high dimensional and distributed nature of a network's internal representations. Interactive demo and code available at https://ndey96.github.io/neuron-explanations-sacrifice.

cs.LG↗