Douglas Bemis

Title: Innovations of Douglas Bemis

Introduction

Douglas Bemis is an accomplished inventor based in San Francisco, CA. He has made significant contributions to the field of neural networks, holding a total of three patents. His work focuses on enhancing the capabilities of computer models through innovative techniques.

Latest Patents

One of his latest patents is titled "Intelligent Regularization of Neural Network Architectures." This invention involves a trained computer model that includes both a direct network and an indirect network. The indirect network generates expected weights or an expected weight distribution for the nodes and layers of the direct network. These expected characteristics can be utilized to regularize the training of the direct network weights, encouraging them towards the expected values predicted by the indirect network. Additionally, the expected weight distribution can be employed to probabilistically predict the output of the direct network based on the likelihood of various weights or weight sets provided by the expected distribution. The output is generated by sampling weight sets from this distribution and evaluating the sampled sets.

Career Highlights

Douglas Bemis is currently associated with Uber Technologies, Inc., where he applies his expertise in neural networks to advance the company's technological capabilities. His innovative approach has positioned him as a key player in the development of cutting-edge solutions in the tech industry.

Collaborations

Throughout his career, Douglas has collaborated with notable professionals in the field, including Zoubin Ghahramani and Theofanis Karaletsos. These collaborations have further enriched his work and contributed to the advancement of neural network technologies.

Conclusion

Douglas Bemis is a prominent inventor whose work in neural networks has led to significant advancements in the field. His innovative patents and collaborations reflect his commitment to pushing the boundaries of technology.