The A*STAR Artificial Intelligence (AI) Initiative is seeking an exceptional CV/ML Scientist to expand our dynamic and multi-disciplinary AI team. Our current team members have degrees in computer science, information systems, psychology, cognitive science and neuroscience from Berkeley, CMU, EFPL, KU Leuven, NTU, NUS, MIT, Stanford, SUTD, UIUC, etc. We work on some of the most crucial, incredibly hard, and future-oriented problems in AI.
The goal of our programme is to develop next-generation AI technologies that will enable machines to:
1) Learn like humans (as would a child from instruction, demonstration, or experience with a small number of examples);
2) Reason for humans (by being able to explain their reasoning, and to personalise recommendations based on their understanding of individual users); and,
3) Deeply understand humans using knowledge of human needs/motivations, social/cultural norms, actions/intentions and commonsense.
In particular, we are looking for an excellent research scientist with a background in Computer Vision and Machine learning to work on human action anticipation and forecasting. Specifically, this project aims to develop new machine learning models that learn to anticipate human actions. The scientist has to work independently to make scientific progress in human action anticipation and forecasting in both theoretical and application aspects. Prior work on human action recognition, generative adversarial networks, and recurrent neural networks is a plus point.
PhD degree in Computer Vision or Machine Learning with a strong track record in CVPR, ECCV, ICCV, NIPS, ICML, WACV, ICLR, AAAI, IJCAI or journal papers in IEEE TPAMI, IJCV, Pattern Recognition, IEEE TIP and related. Entry level candidates with a track record of the above mentioned are encouraged to apply
Strong programming skills in python (Pytorch, Tensorflow)
Excellent computer vision and machine learning skills are required.