AI Institute at UofSC - #AIISC Courses

Real-time Learning Embedding Next Gen Battery Management Systems for EV Safety

2 ratings

Explore advanced Battery Management Systems for EVs that combine physics-driven models with real-time learning to detect faults, monitor health, and enhance safety, addressing current limitations in preventing battery fires.

• YouTube
• 41 minutes
• Self-Paced
• Free Video

Custom, Targeted, Compact, and Neuro-symbolic AI Models for Health Applications

Explore the development of trustworthy AI models for healthcare beyond generative AI, focusing on neurosymbolic, targeted, and compact approaches that ensure explainability, safety, and guideline adherence for personalized health applications.

• YouTube
• 47 minutes
• Self-Paced
• Free Video

Custom, Compact, and Composite AI Systems

Explore the evolution of AI systems from monolithic models to "compound" systems and "agentic" frameworks, focusing on the Custom, Compact, and Composite AI with Neurosymbolic approach for building robust, intelligent AI.

• YouTube
• 1 hour 2 minutes
• Self-Paced
• Free Video

Reconciling Accuracy, Cost, and Latency of Inference Serving Systems

Explore the complex balance between accuracy, cost, and latency in AI inference serving systems with Professor Pooyan Jamshidi's comprehensive analysis.

• YouTube
• 1 hour 21 minutes
• Self-Paced
• Free Video

Edge-Centric LLMs: Optimizing for High-Throughput on Embedded Devices

Explore optimization techniques for deploying Large Language Models on embedded devices, focusing on high-throughput performance with Prof. Ramtin Zand.

• YouTube
• 44 minutes
• Self-Paced
• Free Video

AI For Collaborative Autonomy

Explore collaborative autonomy in AI systems with Dr. Vignesh Narayanan, focusing on multi-agent coordination, decision-making, and real-world applications.

• YouTube
• 51 minutes
• Self-Paced
• Free Video

AI Reasoning and Abstraction

Explore the frontier of AI reasoning and abstraction capabilities with Prof. Amit Sheth's keynote examining whether AI can truly reason like humans, part of the "Beyond the hype" webinar series.

• YouTube
• 25 minutes
• Self-Paced
• Free Video

LLM Hallucinations: Characterization, Quantification, Detection, Avoidance, Mitigation

Explore the complex issue of LLM hallucinations with Vipula Rawte, covering characterization, measurement, detection methods, and strategies for prevention and mitigation.

• YouTube
• 1 hour 13 minutes
• Self-Paced
• Free Video

Explainable Process Recommendation by Contextual Grounded Dynamic Multimodal Process KG

Explore a neurosymbolic framework using Dynamic Multimodal Process Knowledge Graphs for explainable process recommendations, demonstrated through recipe analysis and AI pipeline recommendation use cases.

• YouTube
• 1 hour 20 minutes
• Self-Paced
• Free Video

SmartPilot: A Multiagent CoPilot for Adaptive and Intelligent Manufacturing

Explore SmartPilot, a neurosymbolic multi-agent system that integrates anomaly prediction, demand forecasting, causal reasoning, and real-time decision support to revolutionize adaptive manufacturing with explainable AI.

• YouTube
• 19 minutes
• Self-Paced
• Free Video

Multiagent CoPilot in Industrial AI Applications

Explore the Multiagent CoPilot approach for industrial AI, combining knowledge graphs, rule-based systems, and neurosymbolic AI to create explainable solutions for Industry 4.0 challenges with practical development insights.

• YouTube
• 2 hours 47 minutes
• Self-Paced
• Free Video

Neurosymbolic Customized and Compact Copilots for Critical Applications

Explore the development of neurosymbolic AI models for critical applications, focusing on creating controllable, compact, and customizable systems that ensure reliable and grounded reasoning.

• YouTube
• 1 hour 21 minutes
• Self-Paced
• Free Video

Custom, Targeted, and Neurosymbolic AI Models for Healthcare Applications

Explore the critical role of specialized AI models in healthcare, focusing on neurosymbolic approaches and compact solutions for improved medical applications and outcomes.

• YouTube
• 44 minutes
• Self-Paced
• Free Video

Towards Infusing Auxiliary Knowledge for Distracted Driver Detection

Discover how auxiliary knowledge and scene graphs enhance distracted driver detection systems, improving accuracy by 13.64% through innovative integration of visual cues and driver pose information.

• YouTube
• 16 minutes
• Self-Paced
• Free Video

Evaluating and Rating AI Systems for Trust through a Causal Lens

Explore how causal analysis can enhance AI system evaluation and trust ratings, examining methodologies for more reliable and transparent artificial intelligence assessment.

• YouTube
• 38 minutes
• Self-Paced
• Free Video