Enhanced Suite of Probabilistic Models for Benchmarking Agentic Debugging Frameworks in Probabilistic Inference Article Swipe

MAKE IMPACT

Shibah, Sami Rashid Mohammed · YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.5281/zenodo.17869519

This paper presents an extension to the foundational Reactive Oxygen Species (ROS) stochastic differential equation (SDE) model through four probabilistic models implemented in PyMC v5.25.0+. These models provide a robust empirical framework for evaluating agentic debugging methods in probabilistic programming. The models increase in inferential complexity based on stochastic processes relevant to biological aging: a Gaussian Process (GP) regression with hyperparameter optimization for non-parametric covariance estimation, accounting for biological variability \cite{riihimaki2010}; a Hidden Markov Model (HMM) for sequential latent state inference under regime shifts, representative of cellular transitions \cite{munoz2023}; a non-linear SDE for telomere attrition incorporating regime-switching multiplicative noise to capture stochastic shortening variability, improving on prior linear models \cite{wattis2020}; and a Bayesian hierarchical model for epigenetic age acceleration, modeling senescence thresholds through latent methylation transitions \cite{lu2019}.Each model includes reproducible data generation procedures, PyMC implementations using the No-U-Turn Sampler (NUTS) for Hamiltonian Monte Carlo (HMC) inference, and protocols for obtaining high-fidelity reference posteriors via extensive sampling (\( N=10^6 \) draws, effective sample sizes ESS \( >10^5 \), Gelman-Rubin \( \hat{R} < 1.001 \)). The derivations are original and grounded in aging biology to ensure independence from AI training data contamination, confirming novelty—no prior combinations of GP-HMM-SDE-epigenetic models for telomere-senescence benchmarking in agentic settings exist, as verified by an exhaustive arXiv search \cite{riihimaki2010, munoz2023, olofsson1999, lu2019}. Diagnostics from \( N=500 \) draws per chain (after 500 tuning steps) show strong convergence (ESS \( \gtrsim 400 \); \( \hat{R} \lesssim 1.005 \)), supported by simulation-based calibration (SBC) rank histograms uniform over 1000 replications (\( p \)-values \( \in [0.05, 0.95] \)) and prior sensitivity analyses with \( <3\% \) shifts in posterior means.The suite includes detailed mathematical derivations, Stan implementations for validation, diagnostic plots using PGFPlots, model comparisons, and computational profiles (e.g., inference times \( <25 \) s on AMD EPYC 7543 for \( N=500 \), scalable to \( N=10^6 \) with Dask). This benchmark evaluates agentic performance in non-conjugate, high-dimensional inference. Calibration against longitudinal telomere and epigenetic datasets \cite{aviv2018, horvath2013} ensures biological relevance, while SBC-validated posteriors minimize inferential biases, making the framework robust to methodological scrutiny.

Type

dataset

Landing Page

https://doi.org/10.5281/zenodo.17869519

OA Status

green

OpenAlex ID

https://openalex.org/W7111162947

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W7111162947

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.5281/zenodo.17869519

Digital Object Identifier

Title

Enhanced Suite of Probabilistic Models for Benchmarking Agentic Debugging Frameworks in Probabilistic Inference

Work title

Type

dataset

OpenAlex work type

Publication year

2025

Year of publication

Publication date

2025-12-09

Full publication date if available

Authors

Shibah, Sami Rashid Mohammed

List of authors in order

Landing page

https://doi.org/10.5281/zenodo.17869519

Publisher landing page

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://doi.org/10.5281/zenodo.17869519

Direct OA link when available

Concepts

Computer science, Probabilistic logic, Machine learning, Markov chain Monte Carlo, Statistical model, Model selection, Gaussian process, Algorithm, Bayesian inference, Linear model, Probabilistic relevance model, Artificial intelligence, Gibbs sampling, Inference, Bayesian probability, Covariance, Markov chain, Benchmarking, Statistical inference, Hyperparameter, Data mining, Frequentist inference, Theoretical computer science, Bayes' theorem, Data modeling

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Full payload