Skip to content
VirtualOnlinePast event

Advances in AI at Johns Hopkins University

Event artwork for Advances in AI at Johns Hopkins University

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events
About this event

What to expect

Explore cutting-edge artificial intelligence research from Johns Hopkins University in this online meetup featuring four leading researchers presenting their latest work. Taking place on April 23, 2026 at 9AM PST, this virtual event brings together the global AI community to dive deep into advances in computer vision, deep learning, and computational pathology.

Featured Speakers & Topics

Aniket Roy (Research Scientist, NEC Labs America) will present "Recent Advancements in Image Generation and Understanding," tracing the evolution from GANs to diffusion models. Learn about three breakthrough approaches: DiffNat (TMLR'25) for sharper image generation using kurtosis concentration, DuoLoRA (ICCV'25) for efficient content-style personalization with 19× fewer parameters, and Cap2Aug (WACV'25) for semantic augmentation to address data scarcity in few-shot learning scenarios.

Basudha Pal (PhD, Electrical and Computer Engineering, Johns Hopkins) explores "From Representation Analysis to Data Refinement: Understanding Correlations in Deep Models." Discover how mutual information-based techniques reveal how neural networks implicitly encode demographic and anatomical attributes, and how data attribution methods identify influential training samples to improve model robustness, fairness, and reliability in high-stakes applications.

Tawsifur Rahman (PhD candidate, Biomedical Engineering, Johns Hopkins) presents "Scalable & Precise Histopathology: Next-Gen Deep Learning for Digital Histopathology." Address the computational challenge of gigapixel whole slide images through Task-Specific Self-Supervised Learning (TS-SSL) for domain-optimized features and CEMIL, a contextual attention-based framework achieving state-of-the-art cancer classification while using only a fraction of image patches.

The final talk covers trustworthy AI under real-world data challenges, examining how AI models perform when training assumptions about clean, properly annotated, and diverse data don't hold in practice.

Who Should Attend

This event is ideal for AI researchers, machine learning engineers, computer vision practitioners, and anyone working with deep learning models in medical imaging, generative AI, or representation learning. Whether you're tackling few-shot learning challenges, working with high-resolution medical images, or concerned about model fairness and robustness, these talks offer practical insights from peer-reviewed research published at top venues

Running an AI event?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event