Message from Division Chair
Data science and Quantum Information Science division focuses on sharing knowledge, state-of-the-art tools, techniques, research, and emerging industry trends with machine learning applications. Machine learning tools are becoming prevalent with a growing impact across industries. Reflecting the increased relevance, this division aims to create a forum to exchange experiences and discussions on research topics and inspire collaborations, bringing people together from academia to industries.
Important Links
View Presentation Schedule
Conference Timeline
Invited Speaker
AI and Robotics for Specialty Crops
AI and Robotics have been and will continue to play a key role in reducing farming inputs such as labor, water, and fertilizer and increasing productivity and produce quality. Modular sensing, automation and robotics technologies developed in recent years (including mobile device-based Applications), decreasing cost and increasing capabilities of sensing, control and automation technologies such as UAVs, robust AI tools such as deep learning, and increasing emphasis by governments around the world in advancing AI-empowered smart and automated technologies have created a conductive environment to develop and adopt smart, robotic farming systems for the benefit of agricultural industries around the world with a wide range of farming scale and environment. In this presentation, the author will first discuss the importance of AI-empowered precision and automated/robotic systems for the future of farming (Smart Farming, Ag 4.0). He will then summarize past efforts and current status of agricultural automation and robotics in fruit crops. For example, his work on apple harvesting robots achieved a picking rate of ~80% of apples in modern orchards, taking about ~5.0 sec per fruit. His effort on robotic pollination of apple flowers has achieved a pollination success rate of 84% with a cycle time of 4.2 s. The presentation will conclude with an introduction of the novel robotic systems being developed in his program, and discussion on major challenges and opportunities in AI and robotics in agriculture and related areas including future directions in research and development.
Keywords: AI in Agriculture, Robotics, Smart Farming, Precision Agriculture, Fruit Orchards
Division Schedule
Please look below for detailed schedule.
|
Date/Time: |
Abstract Number: ANPA2025-N00069 Presenting Author: Manoj Karkee (Invited) Co-Authors: nan Presenter's Affiliation: Washington State University, USA Title: AI and Robotics for Specialty Crops Location: Virtual Presentation Show/Hide Abstract AI and Robotics have been and will continue to play a key role in reducing farming inputs such as labor, water, and fertilizer and increasing productivity and produce quality. Modular sensing, automation and robotics technologies developed in recent years (including mobile device-based Applications), decreasing cost and increasing capabilities of sensing, control and automation technologies such as UAVs, robust AI tools such as deep learning, and increasing emphasis by governments around the world in advancing AI-empowered smart and automated technologies have created a conductive environment to develop and adopt smart, robotic farming systems for the benefit of agricultural industries around the world with a wide range of farming scale and environment. In this presentation, the author will first discuss the importance of AI-empowered precision and automated/robotic systems for the future of farming (Smart Farming, Ag 4.0). He will then summarize past efforts and current status of agricultural automation and robotics in fruit crops. For example, his work on apple harvesting robots achieved a picking rate of ~80% of apples in modern orchards, taking about ~5.0 sec per fruit. His effort on robotic pollination of apple flowers has achieved a pollination success rate of 84% with a cycle time of 4.2 s. The presentation will conclude with an introduction of the novel robotic systems being developed in his program, and discussion on major challenges and opportunities in AI and robotics in agriculture and related areas including future directions in research and development.
|
||||||
|
Date/Time: |
Abstract Number: ANPA2025-N00070 Presenting Author: Sarina Gautam Co-Authors: Chandra Adhikari Presenter's Affiliation: John Griffin Middle School, Fayetteville, NC 28304, USA Title: ChatGPT: What is it, and how and why does it work? Location: Virtual Presentation Show/Hide Abstract A super-advanced artificial intelligence chatbot based on the generative pre-trained transformer (GPT) architecture, ChatGPT, is widely used because of its reliable and creative responses for prompts submitted to it. To learn patterns and make the machine learned a prediction of the most likely used token, ChatGPT was trained on a considerable amount of text before it was released, and it is constantly improving. There are some machinery limitations to prediction, as it does not think, use common sense, understand like humans, make an educated guess, or use intuition. Consequently, it makes mistakes or provides hallucinated responses. This study concentrates on the big picture of ChatGPT and how and why it works, emphasizing that it predicts the next token based on context one at a time, using a probabilistic statistical model.
Reference: Stephen Wolfram, “What Is ChatGPT Doing … and Why Does It Work?â€, (Wolfram Media, 2023).
|
||||||
|
Date/Time: |
Abstract Number: ANPA2025-N00071 Presenting Author: Shereiff Garrett Co-Authors: Sarina Gautam; Chandra M. Adhikari Presenter's Affiliation: Department of Chemistry, Physics and Materials Science, Fayetteville State University, Fayetteville, NC 28301, USA Title: Machine learning and machine learned prediction in Chest X-ray images: A case study Location: Virtual Presentation Show/Hide Abstract Machine learning and artificial intelligence are fast-growing fields of research in which data is used to train algorithms, learn patterns, and make predictions. This approach helps to solve seemingly intricate problems with significant accuracy without explicit programming by recognizing complex relationships in data. Taking an example of about 5 thousand chest X-ray images, we implement a machine learning algorithm and present our analysis in making machine-learned predictions in predicting patients with ailments.
|
||||||
|
Date/Time: |
Abstract Number: ANPA2025-N00072 Presenting Author: Puskar Chapagain Co-Authors: Nabin Malakar; Dipak Rimal Presenter's Affiliation: Southern Arkansas University Title: From Problem-Solvers to Graders: The Emerging Role of AI in Physics Evaluation Location: Virtual Presentation Show/Hide Abstract As generative AI tools gain traction in classrooms across technical disciplines such as Physics, they present both opportunities and challenges for reimagining problem-solving strategies and assessment methodologies. In our recent study, we evaluated the performance of different Large Language Models (LLMs) in solving high school-level Physics problems from the Higher Secondary Education Board (HSEB) exams of Nepal. We demonstrated that GPT-4o could solve high school-level Physics problems with impressive accuracy among the tested models, achieving up to 90% when evaluated by a university expert. Given the promising ability to generate correct answers, a compelling question emerges: can they also reliably evaluate Physics answers? In this context, we aim to explore the potential of these models to act as autonomous grading agents for Physics responses. Specifically, we assess whether LLMs can reliably evaluate answers to a variety of Physics problems, including conceptual, numerical, and derivation-based questions. We compare the grading outcomes of these models with evaluations made by experienced Physics educators to assess alignment, consistency, and grading fidelity.
Since AI-assisted grading could significantly support faculty by saving time and providing students with prompt feedback, this study explores the capabilities and limitations of AI-assisted grading in physics education. Furthermore, we discuss its implications for academic integrity, scalability in assessment, and the potential for generating personalized feedback, which are topics of growing relevance in physics education and shaping the future of physics assessments.
|
||||||
|
Date/Time: |
Abstract Number: ANPA2025-N00073 Presenting Author: Abhinav Pokharel Co-Authors: Rupisha Dangol Presenter's Affiliation: Tribhuvan University (Amrit Science Campus) Title: Quantum Key Distribution Using BB84 Protocol: A Computational Study of Error Rates Location: In-Person Presentation, CDP Show/Hide Abstract Quantum key distribution (QKD) enables a secure communication between two parties using the principles of Quantum Physics. The BB84 protocol, one of the first protocols in QKD, is effective not just for sharing keys but also for detecting eavesdropping. The paper simulated a computational model of QKD, analyzing error rates under varying conditions of the number of bits (n). Using Python-based simulation, this paper simulated key exchanges for n = 10, n = 100, n = 1000 bits. We also incorporated a 0.02 noise probability to simulate real–world conditions for both with and without eavesdropping. Each scenario was repeated over 1000 trials to obtain an average error rate. Our results show that, without eavesdropping, the average error rate remains close to 2 %, showcasing BB84’s stability against small noise. In the presence of an eavesdropper, the average error rate consistently exceeds 26%. Moreover, the standard deviation of error increases as n decreases, indicating higher variability in smaller datasets. While this study focuses on computational validation, the future work will extend towards building a low-cost experimental setup using basic optical tools.
Keywords: BB84 protocol, computational simulation, eavesdropping detection, noise modeling, QKD
|
||||||