Assistant HES, PhD student CS UNIGE|
Jason Ramapuram joined the DMML team in 2015. He completed his Masters in Electrical Engineering (with a focus on Signal Processing) in 2011 from the University of California, Riverside. Until joining the DMML team Jason worked as a Software Engineer at Qualcomm Inc and later as a Machine Learning Engineer at Viasat Inc. Jason has also done academic internships at Apple AI Research and Rockwell Collins Advanced Technology Center.
Jason is actively working on Lifelong Learning for Generative Models, Variational approaches to allow CNN’s to work over ultra-high dimensional image data (eg: 4k / 8k images) and approaches to incorporate non-differentiable functions without neural network pipelines. He has also dabbled in some research attempting to find failings of neural networks in simple learning scenarios.
Self-Supervised MultiModal Versatile Networks Unpublished
Lifelong generative modeling Journal Article
Neurocomputing, 404 , pp. 381 - 400, 2020, ISSN: 0925-2312, (Code: https://bitbucket.org/dmmlgeneva/lifelonggenerativemodeling).
IEEE WCCI 2020, IEEE 2019, (Code: http://github.com/apple/ml-dab).
BMVC 2019 & Bayesian Deep Learning Workshop Neurips, 2018, (Code: https://github.com/jramapuram/variational_saccading).
Continual learning Workshop NeurIPS 2018, 2018.
Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2018, Dublin, Ireland, September 10-14, 2018, Proceedings, Part II, pp. 177–192, 2018.
BMVC 2018, abs/1807.00126 , 2018, (Code: https://github.com/apple/ml-all-pairs).
Lifelong Generative Modeling Journal Article
Elsevier Neurocomputing 2020, abs/1705.09847 , 2017.