overconfidence

Repository containing code for the paper: Fixing Overconfidence in Dynamic Neural Networks

A library for calibrating and visualizing the Softmax function for probabilistic interpretation.

Shared-bike behavioral-economics simulator — 共享单车实验模拟器. Twenty days of ride/no-ride choices to study overconfidence, payment-plan rationality, and the sunk-cost effect of a prepaid package. Simulated teaching data.

Simulation-based research on investor bias in Python. Biases are injected into synthetic traders at known strengths to test whether the standard estimators of behavioural finance recover them, and which confounds fool them.

AI agent metacognition skills distilled from David Dunning's Self-Insight (2005): calibration, Dunning–Kruger awareness, the outside view, and feedback discipline for Claude Code, Codex, and compatible AI agents.

Safety before certainty: an open-source public-data stress-test showing a structured evidence-sufficiency prompt cuts unsafe overconfident clinical answers across four frontier LLMs (Opus 4.8, GPT-5.5, Gemini 3.5 Flash, Grok 4.3), with matched circularity controls and blinded clinician judge-validation.

Measuring and fixing DNN overconfidence on MNIST with ECE and temperature scaling.

Almost everyone is overconfident, and almost no one keeps score. Weather forecasters and professional superforecasters are well calibrated for exactly one reason: they write predictions down with probabilities and get graded. This app gives you the same loop with zero ceremony.