All the models are trained for 200 epochs with stochastic gradient descent with a batch size = 128, momentum = 0.9, and cosine All the hyperparameters were selected with a small grid search. From epoch 150 to epoch 185 the training error of the chunks with size 128/256 decreases below 0.5%, while for smaller chunk sizes it remains above 5%. Random chunks with sizes larger than 128/256 can fit the training set, thus having the same representational power as the whole network on the training data. For W > 128/256 the test accuracy is decaying approximately with the same law as that of independent networks with the same width (see Figure 1). This picture suggests that for CIFAR100 the size of a clone is 128/256, slightly larger than the size of the clones in CIFAR10.

Recent work has shown that a person's sympathetic arousal can be estimated from facial videos alone using basic signal processing. This opens up new possibilities in the field of telehealth and stress management, providing a non-invasive method to measure stress only using a regular RGB camera. In this paper, we present SympCam, a new 3D convolutional architecture tailored to the task of remote sympathetic arousal prediction. Our model incorporates a temporal attention module (TAM) to enhance the temporal coherence of our sequential data processing capabilities. The predictions from our method improve accuracy metrics of sympathetic arousal in prior work by 48% to a mean correlation of 0.77. We additionally compare our method with common remote photoplethysmography (rPPG) networks and show that they alone cannot accurately predict sympathetic arousal "out-of-the-box". Furthermore, we show that the sympathetic arousal predicted by our method allows detecting physical stress with a balanced accuracy of 90% - an improvement of 61% compared to the rPPG method commonly used in related work, demonstrating the limitations of using rPPG alone. Finally, we contribute a dataset designed explicitly for the task of remote sympathetic arousal prediction. Our dataset contains synchronized face and hand videos of 20 participants from two cameras synchronized with electrodermal activity (EDA) and photoplethysmography (PPG) measurements. We will make this dataset available to the community and use it to evaluate the methods in this paper. To the best of our knowledge, this is the first dataset available to other researchers designed for remote sympathetic arousal prediction.

Weights initialization in deep neural networks have a strong impact on the speed of converge of the learning map. Recent studies have shown that in the case of random initializations, a chaos/order phase transition occur in the space of variances of random weights and biases. Experiments then had shown that large improvements can be made, in terms of the training speed, if a neural network is initialized on values along the critical line of such phase transition. In this contribution, we show evidence that the scaling property exhibited by physical systems at criticality, is also present in untrained feedforward networks with random weights initialization at the critical line. Additionally, we suggest an additional data-resizing symmetry, which is directly inherited from the scaling symmetry at criticality.

Treeging combines the flexible mean structure of regression trees with the covariance-based prediction strategy of kriging into the base learner of an ensemble prediction algorithm. In so doing, it combines the strengths of the two primary types of spatial and space-time prediction models: (1) models with flexible mean structures (often machine learning algorithms) that assume independently distributed data, and (2) kriging or Gaussian Process (GP) prediction models with rich covariance structures but simple mean structures. We investigate the predictive accuracy of treeging across a thorough and widely varied battery of spatial and space-time simulation scenarios, comparing it to ordinary kriging, random forest and ensembles of ordinary kriging base learners. Treeging performs well across the board, whereas kriging suffers when dependence is weak or in the presence of spurious covariates, and random forest suffers when the covariates are less informative. Treeging also outperforms these competitors in predicting atmospheric pollutants (ozone and PM$_{2.5}$) in several case studies. We examine sensitivity to tuning parameters (number of base learners and training data sampling proportion), finding they follow the familiar intuition of their random forest counterparts. We include a discussion of scaleability, noting that any covariance approximation techniques that expedite kriging (GP) may be similarly applied to expedite treeging.