When Does Multi-Task Learning Fail? Quantifying Data Imbalance and Task Independence in Metal Alloy Property Prediction

Multi-task learning (MTL) assumes related material properties share underlying physics that can be leveraged for better predictions. We test this by simultaneously predicting electrical resistivity, Vickers hardness, and amorphous-forming ability using 54,028 alloy samples. We compare single-task models against standard and structured MTL. Results reveal a striking dichotomy: MTL significantly degrades regression performance (resistivity $R^2$: 0.897 $\to$ 0.844; hardness $R^2$: 0.832 $\to$ 0.694, $p < 0.01$) but improves classification (amorphous F1: 0.703 $\to$ 0.744, $p < 0.05$; recall +17%). Analysis shows near-zero inter-task weights, indicating property independence. Regression failure is attributed to negative transfer caused by severe data imbalance (52k vs. 800 samples). We recommend independent models for precise regression, while reserving MTL for classification tasks where recall is critical.

The Color-Clinical Decoupling: Why Perceptual Calibration Fails Clinical Biomarkers in Smartphone Dermatology

Smartphone-based tele-dermatology assumes that colorimetric calibration ensures clinical reliability, yet this remains untested for underrepresented skin phototypes. We investigated whether standard calibration translates to reliable clinical biomarkers using 43,425 images from 965 Korean subjects (Fitzpatrick III-IV) across DSLR, tablet, and smartphone devices. While Linear Color Correction Matrix (CCM) normalization reduced color error by 67-77% -- achieving near-clinical accuracy (Delta E < 2.3) -- this success did not translate to biomarker reliability. We identify a phenomenon termed "color-clinical decoupling": despite perceptual accuracy, the Individual Typology Angle (ITA) showed poor inter-device agreement (ICC = 0.40), while the Melanin Index achieved good agreement (ICC = 0.77). This decoupling is driven by the ITA formula's sensitivity to b* channel noise and is further compounded by anatomical variance. Facial region accounts for 25.2% of color variance -- 3.6x greater than device effects (7.0%) -- challenging the efficacy of single-patch calibration. Our results demonstrate that current colorimetric standards are insufficient for clinical-grade biomarker extraction, necessitating region-aware protocols for mobile dermatology.

Deep Stock: training and trading scheme using deep learning

Despite the efficient market hypothesis, many studies suggest the existence of inefficiencies in the stock market, leading to the development of techniques to gain above-market returns, known as alpha. Systematic trading has undergone significant advances in recent decades, with deep learning emerging as a powerful tool for analyzing and predicting market behavior. In this paper, we propose a model inspired by professional traders that look at stock prices of the previous 600 days and predicts whether the stock price rises or falls by a certain percentage within the next D days. Our model, called DeepStock, uses Resnet's skip connections and logits to increase the probability of a model in a trading scheme. We test our model on both the Korean and US stock markets and achieve a profit of N\% on Korea market, which is M\% above the market return, and profit of A\% on US market, which is B\% above the market return.