A Library for Uncertainty Quantification.

RAVEN is a flexible and multi-purpose probabilistic risk analysis, validation and uncertainty quantification, parameter optimization, model reduction and data knowledge-discovering framework.

[ICCV 2021 Oral] Deep Evidential Action Recognition

[CVPR 2023] Bridging Precision and Confidence: A Train-Time Loss for Calibrating Object Detection

Official code for "On Calibrating Diffusion Probabilistic Models"

A toolbox for the calibration and evaluation of simulation models.

Codebase for "A Consistent and Differentiable Lp Canonical Calibration Error Estimator", published at NeurIPS 2022.

Simulating and Optimising Dynamical Models in Python 3

This is the official PyTorch codebase for the ACL 2023 paper: "What are the Desired Characteristics of Calibration Sets? Identifying Correlates on Long Form Scientific Summarization".

Official code for "Exploring and exploiting model uncertainty for robust visual question answering"

[ICLR'26] Official code for "A-TPT: Angular Diversity Calibration Properties for Test-Time Prompt Tuning of Vision-Language Models"

Calibration of the significant Social Force Parameters in Vissim

A python package for fitting ViennaPS models to experimental data.

Calibration of the monodomain model coupled with the Rogers-McCulloch model for the ionic current: design of a protocol for impulse delivery from an ATP device.

Cross-instrument plankton image classification + calibration + Grad-CAM + simple trait extraction.

RLHF Loop System - Learning project with monitoring dashboard, drift detection, and AI feedback analysis built with Claude's assistance

Calibration and uncertainty quantification for ranking systems.

Production-grade numerical calibration of the Heston stochastic volatility model with diagnostics

Newton–Puiseux for CVNNs: complete toolkit for uncertainty mining, confidence calibration and local symbolic-numeric analysis on ECG (MIT-BIH) and wireless IQ data (RadioML 2016.10A).

Machine learning pipeline for kidney stone risk prediction, featuring calibrated models, interpretability (Permutation Importance + PDPs), and a clean modular architecture for clinical decision support.