Stephane Marouani, country manager at MathWorks ANZ, tells us what new trends will dominate in STEM education and how aspiring employees will be required to adopt new learning methods.
The importance of STEM (Science, Technology, Engineering, Mathematics) skills to business development and the future of work has been widely talked about for a few years now and thanks to Digital Transformation – the use of digital technologies like deep learning and predictive maintenance that is evolving every industry – the idea that STEM workers are going to have to continue learning more skills than ever before is going to be more apparent in the coming years, for sure.
The impact of Digital Transformation revolves around three main elements:
First, there is an emphasis on developing a conceptual understanding of the core subject (theoretical knowledge coupled with practical knowledge). Second, there is a need to focus on inter-disciplinary work – engineers and scientists must be comfortable in solving problems outside their domain too. Finally, the project team should be able to use new-age infrastructure such as cloud tech and AI-based systems efficiently.
To help prepare students and create the workforce of the future, here are five key trends that will help drive how STEM skills are developed, taught and learned such that students and people in the workforce who are looking to upskill are equipped and ready to contribute and solve these challenges.
1. Authentic learning is bolstered by authentic assessment
In recent years, we have seen the rise of authentic assessment, where teachers move towards evaluating each individual student’s abilities to use course work on projects, using the professional tools provided to them. The ability to accurately track students’ progress is helped by interactive tools and technology that increase student engagement to improve learning outcomes. Professors are starting to explore and integrate products like MATLAB Grader, into their curriculum, allowing them to provide students immediate feedback and to automatically grade MATLAB code within their course work, both of which help assess individual student performance.
2. A shift from “learning to code” to “coding to learn”
The concept of computational thinking enables solving problems, designing systems, and understanding human behaviour. Bringing this approach into every aspect of education, however, is what will help create future-ready students – ones who can break down large problems into a series of small, more manageable ones with rules to solve these.
3. Tools driving global education collaboration
Over the last decade, we have seen more students learning online, where they work on projects that solve problems through concepts, principles and technologies that are used in the industry. In 2020, we’ll see more countries continue to move to online learning strategies, as they shift from paper-based curriculum to tackling real problems through projects and global collaboration. With access to global communities and common, cloud-based storage locations like GitHub and MATLAB Drive, teachers now have help with accessing, sharing and versioning code gathered from the global community to create real and effective teaching strategies and course work.
4. Growing demand for ‘”bi-lingual” engineers = Computer Science + X expertise
Rapidly rising demand has pushed AI and data science from fields that historically fell under computer science and statistics – blending AI with cross-engineering concepts. In academia, this new paradigm is being viewed as “bi-lingual” where course offerings for domain education e.g, chemistry, signal processing, electrical design, etc, are merged with computing and AI. Seeing the real demand for graduates with these ‘bi-lingual’ skills, and the projected shortages across countries and industries, academic institutions and industry are once again working to address this gap through support for curriculum, video intros like Deep Learning in 11 Lines of MATLAB Code, etc. STEM students who learn more about these techniques will have the advantage of also knowing how to use these tools and, more importantly, when to and when not to use them.
5. Growth of self-paced learning options
As students consume more news anywhere, they often become enamoured by buzzwords like AI, but instead need to focus on how they can gain competencies in actual techniques that give them a professional boost such as statistics or reinforcement learning. For motivated learners, there are multiple ways to build and brush up skills whether its self-paced online tutorials like Deep Learning Onramp for MATLAB, Simulink and Stateflow Onramps for Simulink or training available from MOOCs like EdX and Coursera. In addition to building knowledge, these courses de-mystify AI and allow engineers to see it’s a variant of what they may already know – some are concepts they may have learned under different names like system identification or computer vision. This helps build confidence and facilitates teamwork as they can use the new vocabulary to identify with the team’s needs while knowing it’s something they know how to do and do it successfully.