Debapriya Maji

Title: Debapriya Maji: Innovator in Deep Learning and Image Classification

Introduction

Debapriya Maji is a prominent inventor based in Bengaluru, India. She has made significant contributions to the field of deep learning and image classification, holding a total of four patents. Her work focuses on enhancing the capabilities of convolutional neural networks (CNNs) through innovative methods and techniques.

Latest Patents

One of her latest patents is titled "Multi-label image classification in a deep learning network." This patent describes a method for multi-label image classification using a CNN. The method involves forming a composite image from multiple clipped images and processing this composite image through the CNN to generate a probability vector for each clipped image. The length of the probability vector corresponds to the number of classes the CNN is designed to classify.

Another notable patent is "Parametric power-of-2 clipping activations for quantization for convolutional neural networks." This patent outlines a method for quantizing data for a CNN using a power-of-2 parametric activation (PACT2) function. The PACT2 function organizes the data into a histogram and discards a portion of the data corresponding to the tail of the histogram. A clipping value is then determined by expanding the remaining data to the nearest power of two. This method allows for quantization either through post-training quantization or quantization-aware training, helping to maintain accuracy comparable to floating-point models.

Career Highlights

Debapriya Maji is currently employed at Texas Instruments Corporation, where she continues to innovate in the field of deep learning. Her work has garnered attention for its practical applications and contributions to advancing technology in image processing.

Collaborations

She collaborates with talented individuals such as Manu Mathew and Soyeb Noormohammed Nagori, contributing to a dynamic and innovative work environment.

Conclusion

Debapriya Maji's contributions to deep learning and image classification through her patents reflect her expertise and commitment to innovation. Her work at Texas Instruments Corporation positions her as a key player in advancing technology in this field.