Matthias Fey is the creator of the Pytorch Geometric library and a postdoctoral researcher in deep learning at TU Dortmund Germany. He is a core contributor to the Open Graph Benchmark dataset initiative in collaboration with Stanford University Professor Jure Leskovec.
00:00 Intro
00:50 Pytorch Geometric Inception  
02:57  Graph NNs vs CNNs, Transformers, RNNs
05:00  Implementation of GNNs as an extension of other ANNs 
08:15  Image Synthesis from Textual Inputs as GNNs 
10:48  Image classification Implementations on augmented Data in GNNs 
13:40  Multimodal Data implementation in GNNs 
16:25  Computational complexity of GNN Models
18:55  GNNAuto Scale Paper, Big Data Scalability
24:39  Open Graph Benchmark Dataset Initiative with Stanford, Jure Leskovec and Large Networks
30:14 PyG in production, Biology, Chemistry and Fraud Detection 
33:10 Solving Cold Start Problem in Recommender Systems using GNNs
38:21 German Football League, Bundesliga & Playing in Best team of Worst League 
41:54  Pytorch Geometric in ICLR and NeurIPS and rise in GNN-based papers 
43:27 Intrusion Detection, Anomaly Detection, and Social Network Monitoring as GNN implementation 
46:10  Raw data conversion to Graph format as Input in PyG
50:00 Boilerplate templates for PyG for Citizen Data Scientists 
53:37 GUI for beginners and Get Started Wizards  
56:43  AutoML for PyG and timeline for Tensorflow Version  
01:02:40 Explainability concerns in PyG and GNNs in general
01:04:40 CSV files in PyG and Structured Data Explainability
01:06:32 Playing Bass, Octoberfest & 99 Red Balloons 
01:09:50 Collaboration with Stanford, OGB & Core Team
01:15:25 Leaderboards on Benchmark Datasets at OGB Website, Arvix Dataset 
01:17:11 Datasets from outside Stanford, Harvard, Facebook etc  
01:19:00  Kaggle vs Self-owned Competition Platform 
01:20:00 Deploying Arvix Model for Recommendation of Papers
01:22:40 Future Directions of Research 
01:26:00 Collaborations, Jurgen Schmidthuber & Combined Research 
01:27:30 Sharing Office with a Dog, 2 Rabbits and How to train Cats

Ver todos los episodios