-
Sparks of Artificial General Intelligence(AGI) in Semiconductor Material Science: Early Explorations into the Next Frontier of Generative AI-Assisted Electron Micrograph Analysis
Authors:
Sakhinana Sagar Srinivas,
Geethan Sannidhi,
Sreeja Gangasani,
Chidaksh Ravuru,
Venkataramana Runkana
Abstract:
Characterizing materials with electron micrographs poses significant challenges for automated labeling due to the complex nature of nanomaterial structures. To address this, we introduce a fully automated, end-to-end pipeline that leverages recent advances in Generative AI. It is designed for analyzing and understanding the microstructures of semiconductor materials with effectiveness comparable t…
▽ More
Characterizing materials with electron micrographs poses significant challenges for automated labeling due to the complex nature of nanomaterial structures. To address this, we introduce a fully automated, end-to-end pipeline that leverages recent advances in Generative AI. It is designed for analyzing and understanding the microstructures of semiconductor materials with effectiveness comparable to that of human experts, contributing to the pursuit of Artificial General Intelligence (AGI) in nanomaterial identification. Our approach utilizes Large MultiModal Models (LMMs) such as GPT-4V, alongside text-to-image models like DALLE-3. We integrate a GPT-4 guided Visual Question Answering (VQA) method to analyze nanomaterial images, generate synthetic nanomaterial images via DALLE-3, and employ in-context learning with few-shot prompting in GPT-4V for accurate nanomaterial identification. Our method surpasses traditional techniques by enhancing the precision of nanomaterial identification and optimizing the process for high-throughput screening.
△ Less
Submitted 17 September, 2024;
originally announced September 2024.
-
Preliminary Investigations of a Multi-Faceted Robust and Synergistic Approach in Semiconductor Electron Micrograph Analysis: Integrating Vision Transformers with Large Language and Multimodal Models
Authors:
Sakhinana Sagar Srinivas,
Geethan Sannidhi,
Sreeja Gangasani,
Chidaksh Ravuru,
Venkataramana Runkana
Abstract:
Characterizing materials using electron micrographs is crucial in areas such as semiconductors and quantum materials. Traditional classification methods falter due to the intricatestructures of these micrographs. This study introduces an innovative architecture that leverages the generative capabilities of zero-shot prompting in Large Language Models (LLMs) such as GPT-4(language only), the predic…
▽ More
Characterizing materials using electron micrographs is crucial in areas such as semiconductors and quantum materials. Traditional classification methods falter due to the intricatestructures of these micrographs. This study introduces an innovative architecture that leverages the generative capabilities of zero-shot prompting in Large Language Models (LLMs) such as GPT-4(language only), the predictive ability of few-shot (in-context) learning in Large Multimodal Models (LMMs) such as GPT-4(V)ision, and fuses knowledge across image based and linguistic insights for accurate nanomaterial category prediction. This comprehensive approach aims to provide a robust solution for the automated nanomaterial identification task in semiconductor manufacturing, blending performance, efficiency, and interpretability. Our method surpasses conventional approaches, offering precise nanomaterial identification and facilitating high-throughput screening.
△ Less
Submitted 24 August, 2024;
originally announced August 2024.
-
Vision HgNN: An Electron-Micrograph is Worth Hypergraph of Hypernodes
Authors:
Sakhinana Sagar Srinivas,
Rajat Kumar Sarkar,
Sreeja Gangasani,
Venkataramana Runkana
Abstract:
Material characterization using electron micrographs is a crucial but challenging task with applications in various fields, such as semiconductors, quantum materials, batteries, etc. The challenges in categorizing electron micrographs include but are not limited to the complexity of patterns, high level of detail, and imbalanced data distribution(long-tail distribution). Existing methods have diff…
▽ More
Material characterization using electron micrographs is a crucial but challenging task with applications in various fields, such as semiconductors, quantum materials, batteries, etc. The challenges in categorizing electron micrographs include but are not limited to the complexity of patterns, high level of detail, and imbalanced data distribution(long-tail distribution). Existing methods have difficulty in modeling the complex relational structure in electron micrographs, hindering their ability to effectively capture the complex relationships between different spatial regions of micrographs. We propose a hypergraph neural network(HgNN) backbone architecture, a conceptually alternative approach, to better model the complex relationships in electron micrographs and improve material characterization accuracy. By utilizing cost-effective GPU hardware, our proposed framework outperforms popular baselines. The results of the ablation studies demonstrate that the proposed framework is effective in achieving state-of-the-art performance on benchmark datasets and efficient in terms of computational and memory requirements for handling large-scale electron micrograph-based datasets.
△ Less
Submitted 21 August, 2024;
originally announced August 2024.
-
The Discrete Hilbert Transform for Non-Periodic Signals
Authors:
Sumanth Kumar Reddy Gangasani
Abstract:
This note investigates the size of the guard band for non-periodic discrete Hilbert transform, which has recently been proposed for data hiding and security applications. It is shown that a guard band equal to the duration of the message is sufficient for a variety of analog signals and is, therefore, likely to be adequate for discrete or digital data.
This note investigates the size of the guard band for non-periodic discrete Hilbert transform, which has recently been proposed for data hiding and security applications. It is shown that a guard band equal to the duration of the message is sufficient for a variety of analog signals and is, therefore, likely to be adequate for discrete or digital data.
△ Less
Submitted 26 February, 2008;
originally announced February 2008.
-
Analysis of Prime Reciprocal Sequences in Base 10
Authors:
Sumanth Kumar Reddy Gangasani
Abstract:
Prime reciprocals have applications in coding and cryptography and for generation of random sequences. This paper investigates the structural redundancy of prime reciprocals in base 10 in a manner that parallels an earlier study for binary prime reciprocals. Several different kinds of structural relationships amongst the digits in reciprocal sequences are classified with respect to the digit in…
▽ More
Prime reciprocals have applications in coding and cryptography and for generation of random sequences. This paper investigates the structural redundancy of prime reciprocals in base 10 in a manner that parallels an earlier study for binary prime reciprocals. Several different kinds of structural relationships amongst the digits in reciprocal sequences are classified with respect to the digit in the least significant place of the prime. It is also shown that the frequency of digit 0 exceeds that of every other digit when the entire set of prime reciprocal sequences is considered.
△ Less
Submitted 24 January, 2008;
originally announced January 2008.
-
Testing Kak's Conjecture on Binary Reciprocal of Primes and Cryptographic Applications
Authors:
Sumanth Kumar Reddy Gangasani
Abstract:
This note considers reciprocal of primes in binary representation and shows that the conjecture that 0s exceed 1s in most cases continues to hold for primes less one million. The conjecture has also been tested for ternary representation with similar results. Some applications of this result to cryptography are discussed.
This note considers reciprocal of primes in binary representation and shows that the conjecture that 0s exceed 1s in most cases continues to hold for primes less one million. The conjecture has also been tested for ternary representation with similar results. Some applications of this result to cryptography are discussed.
△ Less
Submitted 2 January, 2008; v1 submitted 12 November, 2007;
originally announced November 2007.
-
Testing D-Sequences for their Randomness
Authors:
Sumanth Kumar Reddy Gangasani
Abstract:
This paper examines the randomness of d-sequences, which are decimal sequences to an arbitrary base. Our motivation is to check their suitability for application to cryptography, spread-spectrum systems and use as pseudorandom sequence.
This paper examines the randomness of d-sequences, which are decimal sequences to an arbitrary base. Our motivation is to check their suitability for application to cryptography, spread-spectrum systems and use as pseudorandom sequence.
△ Less
Submitted 19 October, 2007;
originally announced October 2007.
