Dennis Decoste

Title: Innovations of Dennis Decoste in Image Classification

Introduction

Dennis Decoste is an accomplished inventor based in Palo Alto, CA. He has made significant contributions to the field of image classification through his innovative work on hierarchical deep convolutional neural networks. His expertise in this area has led to advancements that improve existing convolutional neural network technology.

Latest Patents

Dennis Decoste holds 1 patent for his invention titled "Hierarchical deep convolutional neural network for image classification." This patent introduces hierarchical branching deep convolutional neural networks (HD-CNNs) that enhance the capabilities of traditional convolutional neural networks (CNNs). In a HD-CNN, easily distinguishable classes are classified in a higher layer coarse category CNN, while more challenging classifications are handled by lower layer fine category CNNs. The training of HD-CNNs may utilize multinomial logistic loss and a novel temporal sparsity penalty, allowing each branching component to address distinct subsets of categories.

Career Highlights

Dennis Decoste is currently employed at eBay, Incorporated, where he applies his knowledge and skills in machine learning and image processing. His work at eBay has allowed him to contribute to the development of advanced technologies that enhance user experience and product categorization.

Collaborations

Throughout his career, Dennis has collaborated with notable colleagues, including Zhicheng Yan and Robinson Piramuthu. These collaborations have fostered an environment of innovation and have led to the development of cutting-edge solutions in the field of image classification.

Conclusion

Dennis Decoste's contributions to hierarchical deep convolutional neural networks represent a significant advancement in image classification technology. His innovative approach continues to influence the field and showcases the importance of collaboration in driving technological progress.