Nilesh Ahuja is a prominent inventor based in Cupertino, CA. He has made significant contributions to the field of neural networks, holding a total of 5 patents. His work focuses on enhancing the capabilities of deep learning systems, particularly in the areas of out-of-distribution detection and variable bitrate compression.

Latest Patents

One of Nilesh's latest patents is titled "Out-of-distribution detection using a neural network." This innovation utilizes features extracted from one or more layers of a trained deep neural network (DNN) to identify out-of-distribution (OOD) data, such as anomalies. The OOD detection process involves transforming a feature output from a layer of the DNN from a high-dimensional feature space to a lower-dimensional space, followed by a reverse transformation back to the higher-dimensional feature space. This results in a reconstructed feature, and a feature reconstruction error is calculated based on the difference between the reconstructed feature and the original feature output from the DNN. Additionally, the process includes calculating a score based on the feature reconstruction error and generating a visual representation of this error.

Another notable patent is "Compression for split neural network computing to accommodate varying bitrate." This patent describes various systems and methods for providing variable bitrate compression for split DNN computing. The system is designed to manage a split DNN that operates on a compute system and a second system over a communication network. It accesses a performance metric to determine a split point of the DNN, defining a head portion and a tail portion. The system also determines a bottleneck layer configuration for a bottleneck layer at the split point, which includes a bottleneck encoder and decoder. The head portion of the DNN and the bottleneck encoder are executed on the compute system, with the system recurrently accessing updated performance metrics to refine the split point or bottleneck layer configuration.

Career Highlights

Nilesh Ahuja is currently employed at Intel Corporation, where he continues to push the boundaries of technology through his innovative work. His expertise in neural networks has positioned him as a key player in the development of advanced computing solutions.

Collaborations

Nilesh has collaborated with notable colleagues, including Omesh Tickoo and Ibrahima J Ndiour. Their combined efforts contribute to the advancement of technologies in the field

This text is generated by artificial intelligence and may not be accurate.

Please report any incorrect information to [email protected]