Tag: thinking

Ever Heard of Machine Teaching?

 

This article is part of our new Learning research and innovation series, offered by Coorpacademy in association with the EPFL’s (Federal Institute of Technology of Lausanne, Switzerland) LEARN Center. The author is Prof. Pierre Dillenbourg, Professor at the EPFL, Head of the CHILI Lab (Computer-Human Interaction for Learning & Instruction) and Director of the Swiss EdTech Collider.

The terms Machine Learning, Deep Learning, and Artificial Intelligence are on everyone’s lips. But what if we extended this list to something we call ‘Machine Teaching’ – and then speculate on what it might mean for education?

Towards ‘Machine Teaching’

Let’s imagine an algorithm that needs to learn how to identify elephants in pictures. In supervised Machine Learning, it gets an example – e.g. picture-3465 – and a label, such as ‘elephant’ or ‘non-elephant’. Picture-3465 may just be the next in a set of thousands of labelled pictures. But if the 3,464 previous pictures were all of African elephants, the system would learn less from yet another African elephant picture, than if an Asian elephant picture was introduced for the first time.

Similarly, if all the previous pictures showed mostly mature elephants, it would be better for the algorithm’s training to select a younger one. Again, if most of them were side on pictures, a frontal view would improve the knowledge acquired by the algorithm.

In other words, if the examples were not fed to the learning algorithm randomly, but strategically selected, one could optimize the machine’s overall learning performance. In a classroom setting, selecting examples is the role of the teacher: she knows that if all examples of squares given to learners are in a horizontal position, learners will logically infer that a square with a 45 degree rotation is not a square.

Any algorithm that determines the optimal sequence of examples such that they are diverse and sufficiently dissimilar from what has been shown previously to a Machine Learning system can be called a Machine Teaching algorithm.

Why Should We Care about Machine Teaching?

If an algorithm receives random examples as inputs, with no strategic consideration of the type of example and what the algorithm will go on to learn from exposure to this example, then clearly problems will arise. First, we should not confuse the size of the sample data with its intrinsic usefulness: merely feeding big data to a Machine Learning algorithm is not enough to guarantee the AI has learnt well and will perform well in its tasks. Secondly, the algorithm could tend towards taking wrong or biased decisions. Let’s reuse the above example of the identification of elephants from pictures: if the only pictures labeled as “non-elephant” are pictures of white animals, the algorithm might infer that only white animals are to be categorised as non-elephants. Sounds silly, but this kind of biases creep in, and matter. Biased algorithms can reinforce gender stereotypes (as was the case in Google’s translation service), or might suggest wrong decisions about humans (as, for example, decision support systems for judges which over-estimated the probability of recidivism for African-American people).

How Does All This Apply to Education?

The impact of AI on education spreads over three layers: (1) Method: AI may enhance the effectiveness of learning technologies where it is expected to enable a fine adaptation of instruction to individual learner needs: over time, a system may learn which learning activity is optimal for a certain learner profile. (2) Content: AI is changing what students should learn or should not learn and is also accelerating the production of learning material, for instance generating questions from Wikipedia. (3) Management: AI and especially data sciences offer new ways to manage education systems (e.g. predicting students’ failure).

Machine Teaching turns out to be relevant in all of those applications. Personalised learning, based on recommender systems, can only be well adapted to the personal needs of a learner if the data set on which the recommendation is based on is large and equilibrated enough. That means we need non-random data selection in any machine learning, i.e. the algorithm needs to be fed with data on what is effective for all types of learners.

In terms of content, when learning about data science and machine learning, learners need to also learn how to design the optimal dataset that the algorithm will learn from. Engineers are becoming teachers of algorithms by default, because you cannot simply program a Machine Learning algorithm. We need to better facilitate the correct decision-making of the algorithm – the same way a good teacher helps her students to develop problem-solving and critical thinking skills.

Innovation in Learning Science and Educational Technologies are top of our agenda at Coorpacademy, as we see them as critical to our mission to continuously improve the learning experience on our platform, making it even more personalized, flexible and enjoyable for learners.

The author Pierre Dillenbourg

Starting young: learning entrepreneurship

By Lamia Kamal-Chaoui, Director, Centre for Entrepreneurship, SMEs, Regions and Cities, OECD. 

This article is extracted from the White Paper “Get ready for the Skills Economy“. Coorpacademy and Citizen Entrepreneurs, the association constituting the French G20 YEA delegation, co-edited this exclusive collection of insight papers on education, used as a discussion piece for this summit.

You’ll find in the White Paper articles about how building a learning culture can address employability challenges, academic insights on Learning Sciences and computational thinking, or how the content and the container must collide in a Netflix-like way to provide the most personalized Learning experience. Articles are signed by Corporate Learning Leaders from various organizations and institutions: Accenture, BNP Paribas, Coorpacademy, emlyon Business School, EY,  OECD, Swiss Federal Institute of Technology, University of Wyoming…

Starting young: learning entrepreneurship – by Lamia Kamal-Chaoui.

Youth are entrepreneurial! New business creation data across OECD countries for 2012-2016 show that 18-30 year olds were more likely to be working on setting up a new business than their older counterparts (6.6%  versus 6.1%), more likely to be setting up businesses in teams of 3 or more, and had a new business ownership rate matching that of adults of over 30 years old (3.5%) (OECD/ EU, 2017).

However, young people face numerous barriers to entrepreneurship, often over and above those faced by their older peers – in identifying opportunities, accessing financing, developing networks, and managing teams. They also often hesitate to start for fear of failure or because they lack the skills (Figure 1). Entrepreneurship education can be a critical support in helping youth to develop an entrepreneurial spirit and obtain the skills needed to become successful entrepreneurs. It is a high-return investment.

Figure 1: Entrepreneurship skills are a greater barrier to business creation for youth

Percentage of population who responded “yes” to the question:

“Do you have the knowledge and skills to start a business?”, Data from 2012-16

Percentage of population who responded “yes” to the question: “Do you have the knowledge and skills to start a business?”, Data from 2012-16

Notes: See Figure 3.13 in OECD/EU (2017). Source: OECD/EU (2017) using special tabulations of the 2012-16 adult population surveys from the Global Entrepreneurship Monitor (2017).

Efforts are increasing to build entrepreneurship competencies through formal education …

Courses and other supports to build entrepreneurship skills in schools, vocational education and training providers, and higher education institutions have become increasingly common in the last decade. They focus on issues of perception about the desirability and feasibility of the entrepreneurial action – either as an entrepreneur or an entrepreneurial employee – and developing the ability to cope with failure.

“Young people face numerous barriers to entrepreneurship, often over and above those faced by their older peers – in identifying opportunities, accessing financing, developing networks, and managing teams.”

However, educational science shows us that developing certain attitudes, knowledge and skills is more effective if started with early intervention (Cunha and Heckman, 2010).

In the area of entrepreneurship skills, a change of content, pedagogy, learning outcomes, and assessment strategies can be introduced as the student progresses, with a gradual increase in the extent that a start-up orientation is offered (OECD, 2015). Some countries (e.g. the United States, Ireland, and Denmark) have already introduced such a progressive approach, but in most OECD countries there is still a need for more entrepreneurship education activities at lower levels of education (GEM, 2017).

Spotlight on higher education

Higher education institutions (HEIs) can be great generators of entrepreneurial individuals. To do so, they themselves need to adopt entrepreneurial approaches to entrepreneurship teaching and supporting graduates who are motivated to start up new ventures — particularly with half of young people accessing higher education across the OECD area. According to the Global University Entrepreneurial Spirit Students’ Survey across 50 countries in 2016, 8% of students intended to start a business right after graduation and 30% considered this a likely career option five years after graduation. The OECD and European Commission have developed the HEInnovate guiding framework for HEIs in this area (www.heinnovate.eu). It identifies many good practices, such as giving students the possibility to document the entrepreneurship competencies they have developed in their studies and extracurricular activities, for example with diploma supplements or other certificates.

What are key areas for government action?

Develop a progressive approach at each stage of the education process. Educa- tional curricula and systems should lay the foundations of an entrepreneurial mind-set at early stages of learning.

Support for teachers. Effective entrepreneurship education requires adequate preparation time for teachers, tailored education material, and guidelines that facilitate the collaboration with external partners (OECD, 2015). In many countries, teacher networks have been formed to provide peer support (e.g. the U.S. Network for Teaching Entrepreneurship, NFTE).

Closing gaps in start-up support. Start-up support should be provided for students who are motivated and able to start a business in the near future. This can be facilitated by creating close connections between education institutions and local business support organisations. Furthermore, higher education students should be supported to combine studies and start-up efforts, for example by receiving a special status similar to sport champions.

References: 

Cunha F. and J. J. Heckman (2010), “Investing in Our Young People”, in Reynolds, A. J. et al., (eds.), Childhood programs and practices in the first decade of life, Cambridge University Press, New York, 381-414.

GEM (2017), Global Entrepreneurship Mo- nitor Report 2016/2017, published online, www.gemconsortium.org.

OECD (2015), From Creativity to Initiative: Building Entrepreneurial Competencies in Schools. A Guidance Note for Policy Makers, published online, http://www.oecd.org/site/entrepreneurship360/blog/guidancenote-policymakers.html

 OECD/EU (2017), The Missing Entrepreneurs 2017: Policies for Inclusive Entrepreneurship, OECD Publishing, Paris, https://doi.org/10.1787/9789264283602-en.

Computational Thinking Will Be Vital For The Future Job Market

 

This piece has been written by Jean-Marc Tassetto, co-founder of Coorpacademy, and originally published in Enterprise Times. To read it in its original form, it’s here!

Computational Thinking is running fast through every avenue of modern business. Jean-Marc Tassetto looks at why this skill is crucial in today’s increasingly data driven organisations

Computational Thinking (CT) is used in the design and analysis of problems and their associated solutions. It is rapidly establishing itself as the literacy of the 21st Century as digital technologies become the core of the workplace.

Business is being disrupted, which will have a huge impact on the employment vista in the coming years, according to the World Economic Forum’s (WEF) 2016 Future of Jobs report. Developments in artificial intelligence (AI), robotics, 3D printing, nanotechnology and biotechnology, amongst others will change the face of the workplace as we know it. This will significantly affect job creation as well as job displacement. On average, by 2020, more than a third of the desired core skills sets for the majority of occupations will be comprised of skills we do not consider crucial at present, according a WEF poll.

With a rapidly evolving job market it is paramount we prepare for future skills requirements and job content at individual, organisation and government levels. As advanced robotics, autonomous transport, AI and machine learning take over, future workforces will need to concentrate on so called ‘soft skills’ – in other words personal attributes such as persuasion, emotional and social intelligence.

Employee skill requirements are evolving

Skillsets will need to be in tune with the digital age. Not looking to address these issues over the coming years could result in large economic cost to businesses, according to the WEF. It isn’t just coding skills that organisations will need. Cloud, analytics, mobility, security, IoT and blockchain will all require the right skills to make them effective. Technical projects rarely happen in a vacuum, so non technical skills will also be important such as leadership, negotiation and communication, together with social and environmental responsibilities.

A learner-centric approach

In response, organisations must continually invest in training that will provide the right skills going forward – that means both technological advancement and soft skills. Training needs to reflect the way people now consume content. Instead of the marathon training sessions of the past, short bursts of training, as needed and always on, are the way forward. They also need to be made available on mobile devices so staff can learn on the move, where and when they want.

Gone are days of rigorous fixed hours, classroom style learning. Instead, to maximise learning and easily measure success, content needs to be placed online in an intuitive virtual learning environment. This way people can take responsibility for their own training and career development and are thoroughly engaged.

Not only computer scientists

And on course content, this is where CT comes in. CT isn’t just for computer scientists, it is a broad, structured way of looking at a problem. It is basically the approach we take when we consider how a computer can help us to solve complex problems. We aren’t just looking at what the computer does in terms of algorithms and abstractions, but also the various strategies that we can implement on digital systems. This involves breaking down problems into various parts as well as designing and using models and defining abstract concepts.

Even if a person doesn’t know how to program or code a computer – being able to think through a problem in a similar, logical manner and come up with a solution in the digital world is paramount. Designing a user journey for a retailer, for example, today requires breaking it down simple steps to put into algorithmic sequences.

Forward-thinking policymakers

CT is so important to enhancing efficiencies and innovation that governments have started to spotlight CT in their re-skilling roadmaps. In the US, the National Research Council, is ahead of the curve, working on CT for the past eight years. The Carnegie-Mellon University has a Microsoft-sponsored Center for Computational Thinking to advance computing research and computational thinking to improve society.

In Europe the Federal Institute of Technology in Lausanne, Switzerland has introduced CT modules. In addition, the Open University is also running introductions to CT, for example. The National University of Singapore has gone a step further and made CT compulsory, regardless of what course they are studying.

21st century business needs CT

CT will be core to future job opportunities. As technology becomes more sophisticated and pervasive we need to understand how to collect data, filter it. We also need to know where to find what we want and how we can use it in decision making. People need to be confident enough to face problems head on and have the ability to work out logical solutions. CT is the flexible tool that provides a consistent and straightforward problem solving technique.

Increasingly we are finding ourselves collaborating with technology. To ensure that people can deal with data in all its increasing complexity, it is imperative that organisations and their staff from the top down are au fait with CT if they are to flourish in the new age of digital intelligence.

This piece has been written by Jean-Marc Tassetto, co-founder of Coorpacademy, and originally published in Enterprise Times. To read it in its original form, it’s here!

How Thinking Like A Computer Will Help Save Our Jobs

 

This piece has been written by Jean-Marc Tassetto, co-founder of Coorpacademy, and originally published on minutehack.com. To read it in its original form, it’s here!

Historically, IT training has focused on coding skills. Now we need to think more like machines as well.

According to Mary Meeker’s much anticipated, just published 2018 technology predictions, you can expect the pace of the disruption of technology on the way we work to just accelerate – not slow down.

But does that mean fewer jobs, as so many fear – or a completely new set of career opportunities?

The evidence of history points to the latter, as the famous Internet trend analyst herself says: ”New technologies have created and displaced jobs historically… Will technology impact jobs differently this time? Perhaps, but it would be inconsistent with history, as new jobs and services plus efficiencies, plus growth typically are created around new technologies.”

And it’s true technology is disrupting the job market. As the World Economic Forum’s 2016 Future of Jobs report and a recent OECD study also found AI (Artificial Intelligence) in particular looks set to take over more and more tasks.

Some authors claim that only as little as 35% of current skills will still be relevant in five years – others say less, and it’s white collar jobs facing automation upheaval this time round, not just blue.

Step forward Computational Thinking

It seems we are on the cusp of a new automation age for sure. And as the robots move into our workplaces, our job roles will adapt – and with it, the skill sets to remain relevant. Everybody will need to have abilities complementary with digital technology.

But not everybody will be in need of hard programming skills: the future will require more than just being to code in Python or deal with malware.

This could mean skills associated with the Cloud, analytics, mobility, security, IoT and blockchain certainly, but there is a growing consensus that, as a culture, we have to introduce a computational/programming-like approach into all of our approaches to work.

This is being formalised around the movement around Computational Thinking (CT), where the focus is not just on the machine but on the human, whose thinking and learning is enhanced by the machine as job roles involve more and more working with computers.

Computational Thinking is basically the approach we take when we consider how a computer can help us to solve complex problems – i.e. algorithms, the way a Machine Learning program can learn from the data it gets, the limits of computation and so on.

But it also shapes what the person involved in the business process does, like preparing a relevant data set for that task, dividing a problem in useful chunks resolvable for a computer, detecting configurations where automation and parallelisation can be introduced, designing digitally, and so on.

What does this look like in the real world? Say you’ve agreed to meet your friends somewhere you’ve never been before. You would probably plan your route before you step out of your house. You might consider the routes available and which route is ‘best’ – this might be the route that is the shortest, the quickest, or the one, which goes past your favourite shop on the way.

You’d then follow the step-by-step directions to get there. In this case, the planning part is like computational thinking, and following the directions is like programming.

With this definition, it’s immediately clear Computational Thinking is not just for computer scientists. Being able to think through a problem in a similar, logical manner and come up with a solution in the digital world is what matters, and what we may all need, and as our professional lives become increasingly automated, CT related skills will grow in importance.

Whether it’s computational contracts, education analytics, computational agriculture or marketing automation, success is going to rely on being able to work fluently with IT, but always to have your eyes on the bigger picture.

Some forward-thinking policymakers are beginning to try and put this digital extension to traditional education on the horizon.

The US, for instance, is among the early adopters of CT, with its National Research Council and US tech university Carnegie-Mellon has its Microsoft-sponsored Center for Computational Thinking that provides seminars, workshops, research activities on computational thinking in any domain of life.

Leading European Higher Education institutions are following suit, like the Federal Institute of Technology in Lausanne, Switzerland, which has been introducing dedicated CT lessons in all entry-level courses across all disciplines.

In the UK, the Open University is also running introductions to CT, while the National University of Singapore has made CT compulsory for higher education students, regardless of what course they are studying. Globally, Google is pushing hard for the democratisation of CT at early years to 12 education globally, providing a variety of teaching material to educators.

The call to action

But what should the world of business be doing about this huge momentous shift? How do firms incorporate CT approaches into their curricula to help their staff? What can we do to help employees successfully transition and acquire these new skills?

First, it’s absolutely key that you insist employees take time out for education and establish continuous learning programmes. To ensure success, you need to get away from the ‘top-down’ approach of old.

The old method of scheduling fixed hours for input needs to be discarded in favour of a learner-chosen model and a virtual learning environment in which all lessons and material are digital and available, 24×7 and increasingly via mobile and in short bursts.

In addition, incorporating gamification and collaboration features will increase employee engagement by activating the joy of competition and the desire for socialisation and exchange.

Employees are also time-poor and required to face rapid changes in their industries and jobs. What they learn must therefore meet their immediate needs and be adjustable to their level.

Asking them questions before any teaching takes place (the flipped pedagogy model) is a great way to pinpoint their level and means they’ll be offered the lessons they need. Finally, this is the foundation of a move towards adaptive learning, in which content and teaching frameworks are customised to the individual.

Such learner-centric approaches work, and can secure user engagement levels of more than 80%. One of our customers, Schneider Electric, places user centricity at the heart of its training efforts: “Individuals are able to self-pace their learning, and we are experimenting with mobile learning as the next frontier in this journey. Digital learning is now a way of life here.” None of this will succeed if employees don’t see the results for them.

According to a Gartner report published in May, “Place the learner’s experience and the solution’s usability at the top of the priority list for any new learning project.”

Training, be it CT-oriented or not, has to be about the learner experience, encouraging employees to develop all their skills to their full potential and to future-proof their careers – and employers need to offer skills like CT if they are to flourish, too.

Embracing a computational thinking mindset will prepare us to meet anything the digital world of the future can throw at us.

Jean-Marc Tassetto is the former CEO of Google France and co-founder of Coorpacademy, a growing force in the provision of user-centric corporate digital learning solutions. 

To read this piece in its original form, it’s here!

MENOU

Voir l'étude de cas