Spencer Young

Spencer Young

Machine Learning Scientist

Posts by Collection

publications

A General Method for Measuring Calibration of Probabilistic Neural Regressors

Published in 3rd Workshop on Uncertainty Reasoning and Quantification in Decision Making (KDD), 2024

As machine learning systems become increasingly integrated into real-world applications, accurately representing uncertainty is crucial for enhancing their robustness and reliability. Neural networks are effective at fitting high-dimensional probability distributions but often suffer from poor calibration, leading to overconfident predictions. In the regression setting, we find that existing metrics for quantifying model calibration, such as Expected Calibration Error (ECE) and Negative Log Likelihood (NLL), introduce bias, require parametric assumptions, and suffer from information theoretic bounds on their estimating power. We propose a new approach using conditional kernel mean embeddings to measure calibration discrepancies without these shortcomings. Preliminary experiments on synthetic data demonstrate the method’s potential, with future work planned for more complex applications.

Recommended citation: Young, S. & Jenkins, P. (2024). "A General Method for Measuring Calibration of Probabilistic Neural Regressors." 3rd Workshop on Uncertainty Reasoning and Quantification in Decision Making (KDD).
Download Paper | Download Bibtex

Fully Heteroscedastic Count Regression with Deep Double Poisson Networks

Published in 42nd International Conference on Machine Learning, 2025

Neural networks capable of accurate, input-conditional uncertainty representation are essential for real-world AI systems. Deep ensembles of Gaussian networks have proven highly effective for continuous regression due to their ability to flexibly represent aleatoric uncertainty via unrestricted heteroscedastic variance, which in turn enables accurate epistemic uncertainty estimation. However, no analogous approach exists for count regression, despite many important applications. To address this gap, we propose the Deep Double Poisson Network (DDPN), a novel neural discrete count regression model that outputs the parameters of the Double Poisson distribution, enabling arbitrarily high or low predictive aleatoric uncertainty for count data and improving epistemic uncertainty estimation when ensembled. We formalize and prove that DDPN exhibits robust regression properties similar to heteroscedastic Gaussian models via learnable loss attenuation, and introduce a simple loss modification to control this behavior. Experiments on diverse datasets demonstrate that DDPN outperforms current baselines in accuracy, calibration, and out-of-distribution detection, establishing a new state-of-the-art in deep count regression.

Recommended citation: Young, S., Jenkins, P., Da, L., Dotson, J., & Wei, H. (2025). "Fully Heteroscedastic Count Regression with Deep Double Poisson Networks." 42nd International Conference on Machine Learning.
Download Paper | Download Slides | Download Bibtex

Assessing the Probabilistic Fit of Neural Regressors via Conditional Congruence

Published in 28th European Conference on Artifical Intelligence (ECAI), 2025

While significant progress has been made in specifying neural networks capable of representing uncertainty, deep networks still often suffer from overconfidence and misaligned predictive distributions. Existing approaches for measuring this misalignment are primarily developed under the framework of calibration, with common metrics such as Expected Calibration Error (ECE). However, calibration can only provide a strictly marginal assessment of probabilistic alignment. Consequently, calibration metrics such as ECE are distribution-wise measures and cannot diagnose the point-wise reliability of individual inputs, which is important for real-world decision-making. We propose a stronger condition, which we term conditional congruence, for assessing probabilistic fit. We also introduce a metric, Conditional Congruence Error (CCE), that uses conditional kernel mean embeddings to estimate the distance, at any point, between the learned predictive distribution and the empirical, conditional distribution in a dataset. We perform several high dimensional regression tasks and show that CCE exhibits four critical properties: correctness, monotonicity, reliability, and robustness.

Recommended citation: Young, S., Edgren, C., Sinema, R., Hall, A., Dong, N. & Jenkins, P. (2025). "Assessing the Probabilistic Fit of Neural Regressors via Conditional Congruence." 28th European Conference on Artifical Intelligence.
Download Paper | Download Bibtex

Learning to Attribute Products to Prices in Retail Shelf Images

Published in KDD (Under Review), 2026

Product-price attribution, the task of retrieving product pricing information directly from an image of a retail display, is critical for monitoring field execution in brick-and-mortar commerce. Poor price compliance, where displayed and point-of-sale prices diverge, can erode consumer trust, reduce revenue, and invite legal risk. Despite growing interest in AI-powered retail compliance solutions, product-price attribution remains underexplored. Existing methods for this task rely on brittle spatial heuristics, rigid shelf-structure assumptions, or high-resolution, close-up imagery that is expensive to obtain. Even state-of-the-art vision-language models (VLMs) struggle with the fine-grained spatial reasoning required. To address these challenges, we present PriceLens, an end-to-end system for product-price attribution that combines off-the-shelf object detection and OCR with PriceNet, a novel transformer-based association model. PriceNet learns to match detected products with price tags by modeling global spatial and semantic context, enabling robust parsing of visually complex retail displays. We introduce the first benchmark dataset for this task and show that PriceLens significantly outperforms heuristic, structural, and VLM baselines on challenging real-world display images. To support further research, we release our dataset to the academic community.

Recommended citation: Young, S., Jenkins, S., Green, L., Miller, K. & Jenkins, P. (2026). "Learning to Attribute Products to Prices in Retail Shelf Images." 32nd SIGKDD Conference on Knowledge Discovery and Data Mining [Under Review].

talks