(By Morna McDermott)
Here it is: DEEP LEARNING
It’s something we can all start following/investigating.
It’s a word… like GRIT, PERSONALIZED LEARNING, CHOICE, and 21st c LEARNING…words that are code for corporate colonization… meet “deep learning.”
About Deeper Learning
In a purely technical (near indecipherable) way, it can be defined: As a set of Machine Learning algorithms which have one or more hidden layers in Neural Networks. Now Deep Learning systems are used for almost any tasks other Machine Learning algorithms are used for. These can be Classification, Dimensionality Reduction, Object Recognition , Clustering etc. Its generally used on high dimensional data like text, images and sound. (Though there are other problems it has been used to solve).
But it also means this:
A 2012 report by the National Research Council concluded that deeper learning competencies—the ability to apply knowledge to new situations—are associated with better life and work outcomes.
Deep learning is also about emotional recognition software. AND also:
“Deep learning is the fastest-growing field in artificial intelligence, helping computers make sense of infinite amounts of data in the form of images, sound, and text. Using multiple levels of neural networks, computers now have the capacity to see, learn, and react to complex situations as well or better than humans. This is leading to a profoundly different way of thinking about your data, your technology, and the products and services you deliver.”
“So what?” you might ask. Terms and phrases and concepts abound in the field of technology, which is transforming at break neck, mind-numbing speed. And its not all bad.
But it’s the relationship to education, profit, control and corporatization that is of concern for those of us fighting for public education and a democratic society.
Another example of techniques used in DEEP LEARNING, “Automatic learning-style detection could play a significant role in the advancement of educational gaming through personalized learning environments. Biometric devices, such as accelerometers and eye-trackers, are now available for use with mobile devices. These provide an opportunity to move toward adaptive mobile gaming environments, giving potential to track learning-styles directly through avatar movement. “
Of course all of this is delivered through friendly sounding student-loving narratives pushed by organizations such as the NEA, as in this article entitled, Helping Students Embrace Deep Learning:
“Just as a video recorder catches a teacher’s classroom behaviors on film, a learning styles inventory captures the intangible and invisible aspects of what a teacher believes (and enacts) about learning. Kolb’s Learning Styles Inventory (LSI) is a valuable guide for teachers seeking to teach effectively, assess learning in meaningful ways, use technology efficiently, and pursue theory-based classroom research.”
Competency based education is the vehicle toward deep learning, “Competency-based education approaches can offer students greater opportunities for deep and personalized learning; as students work toward achieving competency at their own pace.”
Think of “grit and tenacity” when reading how deep learning is defined here:
“Deeper learning prepares students to
- know and master core academic content;
- think critically and solve complex problems;
- work collaboratively;
- communicate effectively; and
- be self-directed and able to incorporate feedback.” Its also not surprisingly tied to emotional learning or “soft skills” or non cognitive domain as it’s known in the edu-tech industry, such as Affectiva: “emotion-aware AI platform built on deep learning”
Then… it gets creepy:
“Affdex delivers discrete and continuous emotion metrics — measured moment-by-moment from a single face or multiple faces simultaneously, in a video or still image. Accuracy and scale is achieved by highly precise emotion classifiers, trained using deep learning and massive amounts of data. This level of refinement provides for robust and trustworthy emotion and other face-related metrics that fuel our norms and predictive analytics.” And they try and make the case that, “In education, emotion analytics can be an early indicator (aka predictive analytics) of student engagement, driving better learning outcomes.” Better outcomes FOR WHO?
Also see Fullan, M. & Langworthy, M. (2014). A Rich Seam: How New Pedagogies Find Deep Learning. London: Pearson.
Whole school districts are jumping on board:
“Natick Public Schools is a suburban school district located southwest of Boston. The school community has about 5,400 students and 600 educators. Grace Magley is the Director of Online learning in Natick, and her office was established to strengthen and deepen the district’s implementation of digital technologies in support of deeper learning and personalization.”
Deep learning is the gateway toward predictive analytics used to predict learning and the uses of biometrics to measure and manage emotional and behavioral outcomes of learning:
“New Pedagogies for Deep Learning project: Global education stakeholders are working together in partnership to address a key education challenge: how educators can design and practice teaching and learning that leads to more successful futures for all students. The partnership aims to articulate and demonstrate how the true potential for learning can be realized through new pedagogies in a technology-rich society.”
Simply apply the technology-specific notion of deep learning (i.e. facial recognition software) to deep learning from a pedagogical standpoint and you can see how “deep learning” will become the way by which corporations (owners of the software and now your personal data) can use game-based systems to manipulate (consumer) social and emotional human behavior. Here are the aspects of DEEP LEARNING outlined:
Character education — honesty, self-regulation and responsibility, perseverance, empathy for contributing to the safety and benefit of others, self-confidence, personal health and well-being, career and life skills. · Citizenship — global knowledge, sensitivity to and respect for other cultures, active involvement in addressing issues of human and environmental sustainability. · Communication — communicate effectively orally, in writing and with a variety of digital tools; listening skills. · Critical thinking and problem solving — think critically to design and manage projects, solve problems, make effective decisions using a variety of digital tools and resources. · Collaboration — work in teams, learn from and contribute to the learning of others, social networking skills, empathy in working with diverse others. · Creativity and imagination — economic and social entrepreneurialism, considering and pursuing novel ideas, and leadership for action. (Fullan, 2013)
This white paper on New Pedagogies for Deep Learning in which the authors state, “deep learning will include the accelerated use of technology” no fewer than ten times, clearly puts it like this: “technology will figure prominently.” Of course it will. The funders of this white paper includes Pearson, Promethean and Microsoft among others.
That is because online delivery of education relies on biometric data and deep learning technologies in order to access the data they so richly need. Finger prints and facial recognition track the online learners “attendance” or the emotional nature of their responses to questions for example.
But wait, there’s more! “A biometric classroom that will track students’ eye movements, monitor their conversations or even measure their smiles. SensorStar Labs co-founder and engineer, Sean Montgomery, believes gleaning information like this – or bio-sensing as he refers to it – from students and giving it to teachers will improve classroom learning. The technology is called EngageSense and uses off-the-shelf webcams to gather the biometric inputs. Then, algorithms repackage the raw data into usable information for teachers, thus giving them additional tools to tailor lesson plans and improve student engagement.”
So the next time to hear someone promoting “deep learning” ask yourself if we aren’t simply stepping into “deep doo doo.”
image from http://deeperlearning4all.org/about-deeper-learning