[NeurIPS Meetup] Practical limitations and new trends in deep learning

On day 5/5 of our NeurIPS Meetup week, we are pleased to present two talks. One is “Practical limitations of today's deep learning in healthcare” by Andrew Ng and the other is a local contribution: “COOT: Cooperative Hierarchical Transformer for Video-Text Representation Learning” by Simon Ging and Mohammadreza Zolfaghari (both from the University of Freiburg).

Our agenda starts at 7:00pm:
Introduction by freiburg.ai and heidelberg.ai hosts
Talk (with interactive chat) about “Cooperative Hierarchical Transformer for Video-Text Representation Learning”
Talk (with interactive chat) by Andrew Ng about “Practical limitations of today's deep learning in healthcare”
Discussion and Q&A session with domain expert

Abstract 1: COOT: Cooperative Hierarchical Transformer for Video-Text Representation Learning

Many real-world video-text tasks involve different levels of granularity, such as frames and words, clip and sentences or videos and paragraphs, each with distinct semantics. In this paper, we propose a Cooperative hierarchical Transformer (COOT) to leverage this hierarchy information and model the interactions between different levels of granularity and different modalities. The method consists of three major components: an attention-aware feature aggregation layer, which leverages the local temporal context (intra-level, e.g., within a clip), a contextual transformer to learn the interactions between low-level and high-level semantics (inter-level, e.g. clip-video, sentence-paragraph), and a cross-modal cycle-consistency loss to connect video and text. The resulting method compares favorably to the state of the art on several benchmarks while having few parameters.

Bio:
Mohammadreza Zolfaghari is a PhD student at Thomas Brox’ Computer Vision Lab at the University of Freiburg. His scientific interests are in the fields of video understanding, human pose estimation, sparse coding, among others. During his PhD, he already produced an impressive list of publications.

Simon Ging is a Master’s student that already has an accepted NeurIPS paper! His research interests are in the fields of Video-Text Learning, Video Captioning and Unsupervised Learning. Before his Masters, he worked as a software developer for SAP systems.

Abstract 2: Andrew Ng: Practical limitations of today's deep learning in healthcare
Recent advances in training deep learning algorithms have demonstrated potential to accommodate the complex variations present in medical data. In this talk, I will describe technical advancements and challenges in the development and clinical application of deep learning algorithms designed to interpret medical images. I will also describe advances and current challenges in the deployment of medical imaging deep learning algorithms into practice.
Bio: Andrew Ng is Founder of DeepLearning.AI, Founder and CEO of Landing AI, General Partner at AI Fund, Chairman and Co-Founder of Coursera, and an Adjunct Professor at Stanford University. As a pioneer both in machine learning and online education, Dr. Ng has changed countless lives through his work and research in the field of artificial intelligence.
Zoom-link to event: https://zoom.us/j/94929450871