ABEMGNIGNI NJIFON Marianne

"Nothing happens until something moves" (Albert Einstein.)

Email: nabemgnigni at aimsammi dot org

About me

I am a PhD Student at the University of Goettingen working on Graph Convolutional Networks and Spatial Inference. I am honoured to be supervised by Pr. Dominic Schuhmacher in the RTG2088 research group. I am interested in developing Graph-based machine learning methods and investigating their mathematical foundations as well as their robustness and Interpretability. I am currently working on applications using Spatial data.

Previously I was an AI Resident in the Google Brain Accra Lab where I worked on frequency analysis to improve the robustness of Deep Neural Networks. I also was a research intern in the Deep Learning Embedded group at DeepMind in London. I was part of the first cohort of the African Master’s of Machine Intelligence in Rwanda. Before that, I completed the AIMS Cameroon Master’s in Mathematical Sciences as a Mastercard Foundation Scholar, where I graduated with honours and as the second best female student. I hold a Master in Physics from the University of Yaounde I.

In addition to Machine Learning research, I am passionate about Physics, especially Nonlinear Optics. I am currenly investigating the propagation of Few-cycle pulses in magnetic and non-magnetic materials with my mentor Pr. Kofane Timoleon Crepin, at the University of Yaounde I . Please check out our paper on this topic that was published in the Journal of Optical Society of America.

Outside of research, I am an advocate for promoting the study of AI and STEM by underrepresented communities, especially in Africa. I am co-lead of Bel’s AI Initiative which conducts outreach about AI in Central Africa with a focus on rural areas. I serve as a mentor to Bel’s AI members. I am part of the WIMLDS Yaounde chapter, and co-organiser of KMERAI events.

If you would like to connect, please reach out at: nabemgnigni at aimsammi dot org

News

I was thrilled to be part of the program comittee at the Neurips 2022 Interpolate workshop, AFCP workshop and BAI workshop.
It was a privilege to attend the wonderful Probai summer school in Helsinki.
Excited to share my interview about my experience with wordsthatcount
It was an honour to present my Physics works as an Invited Speaker at the Third Global Summit on Physics, GSP 2021.
It was a pleasure to give a talk at IndabaX Zimbambwe, 2021
I was thrilled to speak about the Pro’s and Con’s of AI in Healthcare, at the KMERAI workshop, 2021.
Excited to start my PhD with Pr. Dominic Schuhmacher, in the Research training group RTG2088 at the University of Goettingen, Germany.
I just wrapped my AI residency in the Brain team at the Google AI Accra Lab. Thank you to my incredible mentor Yann Dauphin, and to all my colleagues.
It was great spending the past three months in the Deep Learning Embedded group at Deepmind as a Research Intern. I worked on Instance Segmentation and learned a lot from my mentor Aidan Clark.

Publications

Robustmix: Improving Robustness by Regularizing the Frequency Bias of Deep Nets, Jonas Ngnawe(#), Marianne ABEMGNIGNI NJIFON(#), Jonathan Heek, Yann Dauphin, Accepted at the Neurips 2022 workshop on Distribution Shifts: Connecting Methods and Applications. ((#) stands for equal contribution.)
Few-cycle optical pulses in negative index materials with dispersive permittivity and permeability, Marianne ABEMGNIGNI NJIFON, Conrad Bertrand TABI, Timoleon Crepin KOFANE, vol. 37, No. 11, Journal of the Optical Society of America B, 2020.