# Monte Carlo Dropout

2 related articles

Subscribe to updates

arXiv Oct 12, 2026
Desertification forecasts hold for the regional trend, not for individual 2.2 km cells: a 24-year Ceará record puts a number on the ceiling
Over Brazil's semi-arid Ceará state, a study using ten Earth Observation variables from 2000 to 2023 measured how much desertification is actually predictable: the state-wide mean degradation index rose about 0.005 per year, yet most year-to-year change at 2.2 km is not explained by these satellite data, and a gradient-boosting model given every channel explained only 4.7% of its variance. A residual ConvLSTM that predicts only the change from the last observed state lowered RMSE against persistence by 1.6% one year ahead and 4.4% four years ahead on a held-out southern band, reproducing 53–69% of the regional drift. The forecast is reliable for the regional trend, not for individual cells.
Over Brazil's semi-arid Ceará state, a study using ten Earth Observation variables from 2000 to 2023 measured how much desertification is actually predictable: the state-wide mean degradation index rose about 0.005 per year, yet most year-to-year change at 2.2 km is not explained by these satellite data, and a gradient-boosting model given every channel explained only 4.7% of its variance. A residual ConvLSTM that predicts only the change from the last observed state lowered RMSE against persistence by 1.6% one year ahead and 4.4% four years ahead on a held-out southern band, reproducing 53–69% of the regional drift. The forecast is reliable for the regional trend, not for individual cells.
Filipe Moura Da Silva（First author）
Desertification Forecasting Residual ConvLSTM ConvLSTM Earth Observation Dataset Monte Carlo Dropout
arXiv Oct 8, 2026
CoNN-AL reaches R² 0.967–0.982 in automotive glass run channel design using about 2.3% of labels via stream-based active learning
The study proposes CoNN-AL, which adds stream-based active learning to the Cooperative Neural Network with Denoising Autoencoder (CoNN-DAE) and uses Monte Carlo dropout uncertainty to decide in real time which incoming candidates are worth labeling for data-efficient partial inverse design; on a real-world automotive glass run channel dataset of more than 900,000 unique simulated designs, only 20,000 actively selected labels (about 2.3% of the training pool) yield R-squared of 0.967 to 0.982 across all missing-variable levels, approaching upper-bound models trained on far more data, and at harder levels it reaches R-squared of at least 0.95 with 30 to 40% fewer labels than random sampling.
The study proposes CoNN-AL, which adds stream-based active learning to the Cooperative Neural Network with Denoising Autoencoder (CoNN-DAE) and uses Monte Carlo dropout uncertainty to decide in real time which incoming candidates are worth labeling for data-efficient partial inverse design; on a real-world automotive glass run channel dataset of more than 900,000 unique simulated designs, only 20,000 actively selected labels (about 2.3% of the training pool) yield R-squared of 0.967 to 0.982 across all missing-variable levels, approaching upper-bound models trained on far more data, and at harder levels it reaches R-squared of at least 0.95 with 30 to 40% fewer labels than random sampling.
Agung Nugraha（First author）
CoNN-AL Active Learning Partial Inverse Design Cooperative Neural Network with Denoising Autoencoder Monte Carlo Dropout

You’re all caught up

Desertification forecasts hold for the regional trend, not for individual 2.2 km cells: a 24-year Ceará record puts a number on the ceiling

CoNN-AL reaches R² 0.967–0.982 in automotive glass run channel design using about 2.3% of labels via stream-based active learning