Toward Precise n-Type Doping Control in MOVPE-Grown β-Ga2O3 Thin Films by Deep-Learning Approach

In this work, we train a hybrid deep-learning model (fDNN, Forest Deep Neural Network) to predict the doping level measured from the Hall Effect measurement at room temperature and to investigate the doping behavior of Si dopant in both (100) and (010) β-Ga2O3 thin film grown by the metalorganic vapor phase epitaxy (MOVPE). The model reveals that a hidden parameter, the Si supplied per nm (mol/nm), has a dominant influence on the doping process compared with other process parameters. An empirical relation is concluded from this model to estimate the doping level of the grown film with the Si supplied per nm (mol/nm) as the primary variable for both (100) and (010) β-Ga2O3 thin film. The outcome of the work indicates the similarity between the doping behavior of (100) and (010) β-Ga2O3 thin film via MOVPE and the generality of the results to different deposition systems.

The Riccati Equation and the Hidden Parameter &Variable

Using suitable function transformation in combination with a specific Riccati-type equation solvable, general solution of the Riccati equation in the form of elementary quadrature is given. In the process of solving the Riccati equation, the hidden parameter and variable are discovered. This indicates that hidden parameter & variable exist in all differential equations associated with the Riccati equation, such as the second-order linear ODEs, the Schrödinger equation and the Navier–Stokes equation.

Generalized Hidden Parameter MDPs:Transferable Model-Based RL in a Handful of Trials

There is broad interest in creating RL agents that can solve many (related) tasks and adapt to new tasks and environments after initial training. Model-based RL leverages learned surrogate models that describe dynamics and rewards of individual tasks, such that planning in a good surrogate can lead to good control of the true system. Rather than solving each task individually from scratch, hierarchical models can exploit the fact that tasks are often related by (unobserved) causal factors of variation in order to achieve efficient generalization, as in learning how the mass of an item affects the force required to lift it can generalize to previously unobserved masses. We propose Generalized Hidden Parameter MDPs (GHP-MDPs) that describe a family of MDPs where both dynamics and reward can change as a function of hidden parameters that vary across tasks. The GHP-MDP augments model-based RL with latent variables that capture these hidden parameters, facilitating transfer across tasks. We also explore a variant of the model that incorporates explicit latent structure mirroring the causal factors of variation across tasks (for instance: agent properties, environmental factors, and goals). We experimentally demonstrate state-of-the-art performance and sample-efficiency on a new challenging MuJoCo task using reward and dynamics latent spaces, while beating a previous state-of-the-art baseline with > 10× less data. Using test-time inference of the latent variables, our approach generalizes in a single episode to novel combinations of dynamics and reward, and to novel rewards.

Measurement problem and local hidden variables with entangled photons

It is shown that there is no remote action with polarization measurements of photons in singlet state. A model is presented introducing a hidden parameter which determines the polarizer output. This model is able to explain the polarization measurement results with entangled photons. It is not ruled out by Bell’s Theorem.

Fully populated VCM or the hidden parameter

Least-squares estimates are trustworthy with minimal variance if the correct stochastic model is used. Due to computational burden, diagonal models that neglect correlations are preferred to describe the elevation dependency of the variance of GPS observations. In this contribution, an improved stochastic model based on a parametric function to take correlations between GPS phase observations into account is presented. Built on an adapted and flexible Mátern function accounting for spatiotemporal variabilities, its parameters can be fixed thanks to Maximum Likelihood Estimation or chosen apriori to model turbulent tropospheric refractivity fluctuations. In this contribution, we will show in which cases and under which conditions corresponding fully populated variance covariance matrices (VCM) replace the estimation of a tropospheric parameter. For this equivalence “augmented functional versus augmented stochastic model” to hold, the VCM should be made sufficiently largewhich corresponds to computing small batches of observations. A case study with observations from a medium baseline of 80 km divided into batches of 600 s shows improvement of up to 100 mm for the 3Drms when fully populated VCM are used compared with an elevation dependent diagonal model. It confirms the strong potential of such matrices to improve the least-squares solution, particularly when ambiguities are let float.