arXiv subjects

Jonne Pohjankukka

Publications and source records attributed to Jonne Pohjankukka.

4 recordsLinked to original sources

arXiv2026

Stochastic Gradient Optimization with Model-Assisted Sampling

This work addresses the problem of variance in stochastic gradient estimation for machine learning optimization. Deep learning relies on mini-batch methods such as stochastic gradient descent, which approximate full gradients but introduce noise, creating trade-offs between convergence stability, speed, and generalization. Existing methods, including variance reduction techniques (e.g., SVRG and SAG) and adaptive optimizers, aim to mitigate gradient noise but may introduce additional computational overhead. We propose a model-assisted sampling framework that interprets mini-batch gradients through survey sampling theory, treating the dataset as a fixed finite population and gradients as sample-based estimates. Our aim is to bridge machine learning optimization and survey sampling theory by combining their perspectives on sample-based estimation and variance reduction. By incorporating auxiliary gradient-prediction models, we construct more efficient gradient estimators, with uniform sampling arising as a special case when no auxiliary information is used. Our approach integrates easily with existing optimizers, improving efficiency without altering their dynamics. Empirical results on synthetic and six benchmark datasets show performance gains in 71-86% of the experiments, particularly for medium-sized input spaces in our benchmarks. Notably, with momentum-based optimizers such as AdamW, the proposed estimator achieves clearly better generalization in roughly half the training epochs compared to baseline estimator.

cs.LG↗

arXiv2024

Model-assisted survey sampling with Bayesian optimization

Survey sampling plays an important role in the efficient allocation and management of resources. The essence of survey sampling lies in acquiring a sample of data points from a population and subsequently using this sample to estimate the population parameters of the targeted response variable, such as environmental-related metrics or other pertinent factors. Practical limitations imposed on survey sampling necessitate prudent consideration of the number of samples attainable from the study areas, given the constraints of a fixed budget. To this end, researchers are compelled to employ sampling designs that optimize sample allocations to the best of their ability. Generally, probability sampling serves as the preferred method, ensuring an unbiased estimation of population parameters. Evaluating the efficiency of estimators involves assessing their variances and benchmarking them against alternative baseline approaches, such as simple random sampling. In this study, we propose a novel model-assisted unbiased probability sampling method that leverages Bayesian optimization for the determination of sampling designs. As a result, this approach can yield in estimators with more efficient variance outcomes compared to the conventional estimators such as the Horvitz-Thompson. Furthermore, we test the proposed method in a simulation study using an empirical dataset covering plot-level tree volume from central Finland. The results demonstrate statistically significant improved performance for the proposed method when compared to the baseline.

stat.ME↗

arXiv2020

Utilizing remote sensing data in forest inventory sampling via Bayesian optimization

In large-area forest inventories a trade-off between the amount of data to be sampled and the costs of collecting the data is necessary. It is not always possible to have a very large data sample when dealing with sampling-based inventories. It is therefore necessary to optimize the sampling design in order to achieve optimal population parameter estimation. On the contrary, the availability of remote sensing (RS) data correlated with the forest inventory variables is usually much higher. The combination of RS and the sampled field measurement data is often used for improving the forest inventory parameter estimation. In addition, it is also reasonable to study the utilization of RS data in inventory sampling, which can further improve the estimation of forest variables. In this study, we propose a data sampling method based on Bayesian optimization which uses RS data in forest inventory sample selection. The presented method applies the learned functional relationship between the RS and inventory data in new sampling decisions. We evaluate our method by conducting simulated sampling experiments with both synthetic data and measured data from the Aland region in Finland. The proposed method is benchmarked against two baseline methods: simple random sampling and the local pivotal method. The results of the simulated experiments show the best results in terms of MSE values for the proposed method when the functional relationship between RS and inventory data is correctly learned from the available training data.

cs.LG↗

arXiv2020

Estimating the Prediction Performance of Spatial Models via Spatial k-Fold Cross Validation

In machine learning one often assumes the data are independent when evaluating model performance. However, this rarely holds in practise. Geographic information data sets are an example where the data points have stronger dependencies among each other the closer they are geographically. This phenomenon known as spatial autocorrelation (SAC) causes the standard cross validation (CV) methods to produce optimistically biased prediction performance estimates for spatial models, which can result in increased costs and accidents in practical applications. To overcome this problem we propose a modified version of the CV method called spatial k-fold cross validation (SKCV), which provides a useful estimate for model prediction performance without optimistic bias due to SAC. We test SKCV with three real world cases involving open natural data showing that the estimates produced by the ordinary CV are up to 40% more optimistic than those of SKCV. Both regression and classification cases are considered in our experiments. In addition, we will show how the SKCV method can be applied as a criterion for selecting data sampling density for new research area.

stat.AP↗