Augustin Prado

Title: Innovations of Augustin Prado in Neural Network Architecture

Introduction

Augustin Prado is an accomplished inventor based in Lausanne, Switzerland. He has made significant contributions to the field of neural networks, particularly in understanding uncertainties in predictions. His innovative work has led to the development of a unique patent that addresses complex challenges in geophysical property predictions.

Latest Patents

Augustin Prado holds a patent for a "Dual neural network architecture for determining epistemic and aleatoric uncertainties." This neural network system comprises a first neural network that predicts a mean value output and epistemic uncertainty based on input data. Additionally, a second neural network is designed to predict the total uncertainty of the output from the first neural network. The training process for the second neural network involves minimizing a cost function that compares the predicted mean value of a geophysical property with its ground-truth value. The system also includes processors that determine aleatoric uncertainty based on the epistemic uncertainty and total uncertainty of the output.

Career Highlights

Augustin Prado is currently employed at Schlumberger Technology Corporation, where he applies his expertise in neural networks to advance technological solutions in the geophysical sector. His work is instrumental in enhancing the accuracy of predictions related to geological formations.

Collaborations

Augustin collaborates with notable colleagues, including Ravinath Kausik Kadayam Viswanathan and Lalitha Venkataramanan. Their combined efforts contribute to the innovative research and development within their field.

Conclusion

Augustin Prado's contributions to neural network architecture exemplify the intersection of technology and geophysical research. His patent reflects a significant advancement in understanding uncertainties, showcasing his role as a leading inventor in this domain.