Vukosi Marivate Researcher, Council for Scientific and Industrial Research, South Africa |
|
How DID YOU GET INTO machine learning?
From a young age I have had an interest on how machines can be used to solve challenges. What really drove me was understanding how data fit into this picture. That got me to cross paths with Artificial Intelligence and Machine Learning. I chose the route through a BSc and MSc in Electrical Engineering at the University of the Witwatersrand, Johannesburg. My later years in undergrad and MSc were focused in Machine Learning, under the supervision of Prof. Tshilidzi Marwala. I grew an interest in creating agents that learn from experience and as such my PhD focus was on Reinforcement Learning. Specifically I was interested in techniques to make Reinforcement Learning more robust and ways to evaluate algorithms only using pre-collected data. My PhD was under the supervision of Prof. Michael Littman at Rutgers University.
I have always been driven by using Machine Learning to solve challenges with data and as such moved towards Data Science. I took internships at Google and Knewton, looking at challenges of anomaly detection and tracking students in online education materials respectively. I am now a Data Scientist who looks at challenges facing society, Machine Learning is one of the tools under my belt. Now I work with a multidisciplinary team and look at different problems using ML, Natural Language Processing, Statistics, GIS etc.
I have always been driven by using Machine Learning to solve challenges with data and as such moved towards Data Science. I took internships at Google and Knewton, looking at challenges of anomaly detection and tracking students in online education materials respectively. I am now a Data Scientist who looks at challenges facing society, Machine Learning is one of the tools under my belt. Now I work with a multidisciplinary team and look at different problems using ML, Natural Language Processing, Statistics, GIS etc.
WhAT WILL YOU Be teaching?
We will be looking at the fundamental underpinnings that make up Reinforcement Learning. We will go from Reinforcement Learning where one has complete information of the agent, environment and reward. We then will start removing some of this information and formulating the challenges with these removals, mimicking the challenges an agent would face in the real world.
What advice would you give to those getting started in machine/deep learning?
Patience and taking time to understand the underpinnings of approaches and algorithms is key. Having a curious mind about how the world works and looking at how one can improve society with the skills you learn is a powerful combination to make the world a better place.
