Ann Arbor AI, Machine Learning and Computer Vision Meetup

https://voxel51.com/computer-vision-events/ai-machine-learning-computer-vision-meetup-sept-12-2024/

Reducing Hallucinations in ChatGPT and Similar AI Systems

LLMs are prone to producing hallucinations, largely due to their limited content and knowledge base. One of the most widely used techniques to reduce hallucinations is incorporating external knowledge sources. Among these, using knowledge graphs has shown particularly impressive results in enhancing the accuracy and reliability of the results produced by LLMs. In this talk, we will explore what knowledge graphs are, why they are important, and how to utilize the Neo4j graph database to improve the reliability of LLMs.

About the Speaker

Abhimanyu Aryan started in VR industry, then worked as an ML Engineer (Vision) for the Indian Air Force and contributed to Julia’s open-source web ecosystem( mostly Genie). Currently, building an AI stealth startup.

Update: Data-Centric AI Competition on Hugging Face Spaces

Are you ready to challenge the status quo in AI development? Then join Voxel51’s Harpreet Sahota for the latest updates, plus tips and tricks on the first-ever Data-Centric AI competition on Hugging Face Spaces, focusing on the often-overlooked yet crucial aspect of AI: data curation. Learn more about the competition, rules and prizes.

About the Speaker

Harpreet Sahota is a hacker-in-residence and machine learning engineer with a passion for deep learning and generative AI. He’s got a deep interest in RAG, Agents, and Multimodal AI.

It's in the Air Tonight. Sensor Data in RAG

I will do a quick overview of the basics of Vector Databases and Milvus and then dive into a practical example of how to use one as part of an application. I will demonstrate how to consume air quality data and ingest it into Milvus as vectors and scalars. We will then use our vector database of Air Quality readings to feed our LLM and get proper answers to Air Quality questions. I will show you how to all the steps to build a RAG application with Milvus, LangChain, Ollama, Python and Air Quality Reports. Preview the demo on Medium.

About the Speaker

Tim Spann is a Principal Developer Advocate for Zilliz and Milvus. He works with Milvus, Generative AI, HuggingFace, Python, Big Data, IoT, and Edge AI. Tim has over twelve years of experience with the IoT, big data, distributed computing, messaging, machine learning and streaming technologies.

https://voxel51.com/computer-vision-events/visual-ai-in-healthcare-sept-19-2024/

Interpretable AI Models in Radiology

AI methods have reached and even surpassed human-level accuracy in numerous areas of healthcare. However, adoption of these technologies into clinical workflows, where interpretability is of paramount importance, is slower compared to other industries. In this talk, we will present an overview of our research in improving the interpretability of AI models in medical image analysis through counterfactual examples and radiologist gaze data collection.

About the Speaker

Dr. Tasdizen is a Professor in Electrical and Computer Engineering and the Scientific Computing and Imaging (SCI) Institute at the University of Utah. His areas of expertise are medical image analysis and machine learning.

Bridging Species with Pixels: Advancing Comparative Computational AI in Veterinary Oncology

Roughly 50% of dogs over the age of 10 years will develop cancer. Animals are now part of the family, and veterinary medical care now approximates what is available in humans. We are now at a pivotal time where AI platforms and products can expedite clinical discovery and decision - making and accelerate innovation. In this talk, we will provide a high-level overview of comparative AI and the work our team has initiated to evaluate both radiomic and language-based models in veterinary medicine.

About the Speakers

Dr. Christopher Pinard, DVM DVSc DACVIM (Oncology) is the CEO and co-founder of ANI.ML Health Inc., an adjunct professor in the Department of Clinical Studies at the Ontario Veterinary College, University of Guelph, a Medical Oncologist at Lakeshore Animal Health Partners, a Research Fellow at Sunnybrook Research Institute, and a Faculty Affiliate with the Centre for Advancing Responsible and Ethical Artificial Intelligence (CARE-AI) at the University of Guelph. His research focuses on comparative computational oncology and the development of computer vision and language model-based tools for clinical applications.

Dr. Kuan-Chuen Wu builds A.I. products and Engineering solutions via scientific research, technological development, and global teaching. With a Harvard-Stanford education in multi-disciplinary engineering, data science, and business management, he leads multi-functional teams and communities in generative A.I. and predictive A.I. using hardware, software, theory plus ingenuity for societal good.

Deep-Dive: NVIDIA’s VISTA-3D and MedSAM-2 Medical Imaging Models

In this talk, we’ll explore two medical imaging models. First, we’ll dive into NVIDIA’s Versatile Imaging SegmenTation and Annotation (VISTA) model which combines semantic segmentation with interactivity, offering high accuracy and adaptability across diverse anatomical areas for medical imaging. Finally, we’ll explore MedSAM-2, an advanced segmentation model that utilizes Meta’s SAM 2 framework to address both 2D and 3D medical image segmentation tasks.

About the Speaker

Daniel Gural is a seasoned Machine Learning Engineer at Voxel51 with a strong passion for empowering Data Scientists and ML Engineers to unlock the full potential of their data.

Exploring Instance Imbalance in Medical Semantic Segmentation

Current benchmarks in Medical Semantic Segmentation either leave out imbalanced datasets or focus on class imbalance. However, the nature of semantic segmentation shows that it is construed towards the segmentation of objects without differentiating multiple instances within a single class. This leads to the problem of instance imbalance in semantic segmentation. This is quite concerning in the case of medical image segmentation where the size of instances is principal. This talk will focus on a new evaluation metric and analysis of losses particularly to understand instance imbalance in semantic segmentation.

About the Speaker

Soumya Snigdha Kundu is a Ph.D. student at King’s College London. His work is focused on Trustworthy Machine Learning (TML) and its application to Neuro-Oncology.

Register for the Zoom

About the workshop

Want greater visibility into the quality of your computer vision datasets and models? Then join Harpreet Sahota, Hacker in Residence and Machine Learning Engineer at Voxel51, for this free 90-minute, hands-on workshop to learn how to leverage the open source FiftyOne computer vision toolset.

In the first part of the workshop we’ll cover:

• FiftyOne Basics (terms, architecture, installation, and general usage)
• An overview of useful workflows to explore, understand, and curate your data
• How FiftyOne represents and semantically slices unstructured computer vision data

The second half will be a hands-on introduction to FiftyOne, where you will learn how to:

• Load datasets from the FiftyOne Dataset Zoo
• Navigate the FiftyOne App
• Programmatically inspect attributes of a dataset
• Add new sample and custom attributes to a dataset
• Generate and evaluate model predictions
• Save insightful views into the data

Prerequisites are a working knowledge of Python and basic computer vision. All attendees will get access to the tutorials, videos, and code examples used in the workshop.

About the Instructor

Harpreet Sahota is a hacker-in-residence and machine learning engineer with a passion for deep learning and generative AI. He’s got a deep interest in RAG, Agents, and Multimodal AI.

https://voxel51.com/computer-vision-events/ai-machine-learning-computer-vision-meetup-sept-26-2024/

GPUs at Scale - Trials of a GPUaaS Provider

In the rapidly evolving landscape of machine learning, managing large-scale GPU infrastructure has become a critical challenge for AI practitioners. This presentation delves into the trenches of GPU operations. Through these real-world lessons, we explore how GPU-as-a-Service (GPUaaS) solutions address these pain points, offering scalability, flexibility, and cutting-edge hardware access. Join us for a deep dive into the challenges of computation at scale, the lessons learned from hands-on experience, and a discussion of emerging strategies in GPU infrastructure management.

About the Speaker

Mischa van Kesteren is a Solutions Engineer at NexGen Cloud, focused on Hyperstack, their GPUaaS On-Demand platform. With nearly a decade of experience in the field of HPC and AI, Mischa leads complex solution designs and serves as a thought leader in sustainable and secure AI and HPC practices.

Scaling Industrial AI with FiftyOne

Datasets and models are the two pillars of modern machine learning, but connecting the two can be cumbersome and time-consuming. In this hands-on talk, you will learn how FiftyOne Teams simplifies this complexity in Industrial use cases, enabling more effective data-model co-development. By the end of the talk, you will be able to download and visualize datasets with FiftyOne, explore data embeddings, apply anomaly detection, and effortlessly share your datasets with others.

About the Speaker

Daniel Gural is a seasoned Machine Learning Engineer at Voxel51 with a strong passion for empowering Data Scientists and ML Engineers to unlock the full potential of their data.