Neural network improves prediction of wallet reputations in finance
zScore-N: A Neural Network for On-Chain Wallet Reputation Scoring
Artificial Intelligence
Summary
Wallet reputation scores help decide who gets financial benefits and access in decentralized finance systems. The authors found that traditional methods using hand-made formulas are rigid and cannot learn from new data or handle missing information well. They created a neural network called zScore-N that learns to mimic the original formula but does so more accurately and is better at handling incomplete data. This network works efficiently with millions of wallets and can improve reputation scoring over time as more data becomes available.
wallet reputation scoredecentralized financeneural networkmachine learningmissing dataregressiongradient-boosted treesscore calibrationtraining dataallowlist
Authors
Girish G N, Ashutosh Sahoo, Akshay SP, Gurukiran S, Dhanashekar Kandaswamy
Abstract
Wallet reputation scores decide who receives an airdrop, who can borrow, and who enters an allowlist across decentralised finance. They almost always begin as hand-written formulas: compositions of clamped logarithmic, linear and square-root transforms over behavioural features, with every threshold and point award set by hand. Such a formula is readable and deterministic, but it is piecewise and non-differentiable, it cannot improve as data accumulates, and it cannot distinguish a feature that is genuinely zero from one its pipeline failed to capture. We present zScore-N, the neural network that replaced ours in production. The formula served as its teacher: calibrated against 5,208,952 wallets sampled across 2019-2024 and verified to reproduce production output to within 2.3e-13, it supplies unlimited labelled training data at zero label noise. The trained network reproduces it to 0.58 points RMSE on the 1000-point scale (R^2 = 0.99997), against 2.25 for gradient-boosted trees and 28.04 for linear regression on identical features and splits. Trained with missing-value masks against uncorrupted targets, it halves the error that incomplete data introduces: at 10% feature-level missingness the formula drifts 51.4 points from its own complete-data output with a systematic -12.5 point bias, while the network drifts 17.9. The network carries the score at production scale, across a population of millions of wallets spanning six orders of magnitude in size and activity.