Doctoral Student Rocco Sedona

[In Progress] Investigation on Machine and Deep Learning Methods using Large Quantities of Remote Sensing Data
Rocco Sedona, Doctoral Thesis
University of Iceland, School of Engineering and Natural Sciences (SENS), Iceland


Rocco Sedona

Lectures (given by R. Sedona since Ph.D. Studies Start 2019-04-01)

Plenary Lecture or Keynote Address at an International Academic Conference

none

Public Guest Lecture at a University outside of Iceland

none

Lecture at an International Conference

  1. Tutorial: Hands-On Distributed Deep Learning, IEEE International Geoscience and Remote Sensing Symposium (IGARSS) Brussels, Virtual Conference, July 10, 2021, Virtual Conference
    [ EVENT ] [ Slides ~1.12 MB (pdf) ]
  2. Enhancing Large Batch Size Training of Deep Models for Remote Sensing Applications, IEEE International Geoscience and Remote Sensing Symposium (IGARSS)
    Brussels, Virtual Conference, July 15, 2021, Virtual Conference
    [ EVENT ] [ Slides ~1.31 MB (pdf) ]

  3. SCALING UP A MULTISPECTRAL RESNET-50 TO 128 GPUS, paper presentation, 2020 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), September 26 – October 2, 2020, Virtual Symposium
    [ EVENT ]
  4. FD-6 Tutorial: Scalable Machine Learning with High Performance and Cloud Computing, one theoretical lecture, one practical lecture, 2020 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), September 26 – October 2, 2020, Virtual Symposium
    [ Slides (~3,18 MB) ] [ EVENT ]
  5. Using machine learning method to classify polar stratospheric cloud types from Envisat MIPAS observation, presentation, 2020 EGU General Assembly (EGU 2020), May 4 – May 8, 2020, Conline Conference
    [ Slides (~2,67 MB) ] [ EVENT ]

Lecture at a Domestic Conference

none

Lecture at an Academic Symposium, Seminar or Forum for Academic Groups

  1. Parallel and Scalable Machine Learning Research for Earth Science Applications, presentation, PHD Mid-Term Evaluation Seminar, The University of Iceland, Groska, 2021, December 16
    [ Slides ~3.92 MB (pdf) ]

  2. How to run remote sensing data with RESNET-50 on HPC systems at scale using Horovod, presentation, Day 2 BOS Session, Challenges and opportunities with running AI workloads on HPC systems, 2020 Joint Laboratory for Extreme-Scale Computing (JLESC), September 8 – September 10, 2020, Virtual Symposium
    [ EVENT ]
  3. Classification methods for Polar Stratospheric Clouds, presentation, Short Talks Sessoin, 2019 Joint Laboratory for Extreme-Scale Computing (JLESC), April 15 – April 17, 2019, Knoxville, TN, USA
    [ EVENT ]

Poster at an International Conference

none

Poster at a Domestic Conference

none