MIT-deep learning chip Eyeriss
Chip could bring deep learning to mobile devices

via MASSACHUSETTS INSTITUTE OF TECHNOLOGY

In recent years, some of the most exciting advances in artificial intelligence have come courtesy of convolutional neural networks, large virtual networks of simple information-processing units, which are loosely modeled on the anatomy of the human brain.

Neural networks are typically implemented using graphics processing units (GPUs), special-purpose graphics chips found in all computing devices with screens. A mobile GPU, of the type found in a cell phone, might have almost 200 cores, or processing units, making it well suited to simulating a network of distributed processors.

At the International Solid State Circuits Conference in San Francisco this week, MIT researchers presented a new chip designed specifically to implement neural networks. It is 10 times as efficient as a mobile GPU, so it could enable mobile devices to run powerful artificial-intelligence algorithms locally, rather than uploading data to the Internet for processing.

Neural nets were widely studied in the early days of artificial-intelligence research, but by the 1970s, they’d fallen out of favor. In the past decade, however, they’ve enjoyed a revival, under the name “deep learning.”

“Deep learning is useful for many applications, such as object recognition, speech, face detection,” says Vivienne Sze, an assistant professor of electrical engineering at MIT whose group developed the new chip. “Right now, the networks are pretty complex and are mostly run on high-power GPUs. You can imagine that if you can bring that functionality to your cell phone or embedded devices, you could still operate even if you don’t have a Wi-Fi connection. You might also want to process locally for privacy reasons. Processing it on your phone also avoids any transmission latency, so that you can react much faster for certain applications.”

The new chip, which the researchers dubbed “Eyeriss,” could also help usher in the “Internet of things” — the idea that vehicles, appliances, civil-engineering structures, manufacturing equipment, and even livestock would have sensors that report information directly to networked servers, aiding with maintenance and task coordination. With powerful artificial-intelligence algorithms on board, networked devices could make important decisions locally, entrusting only their conclusions, rather than raw personal data, to the Internet. And, of course, onboard neural networks would be useful to battery-powered autonomous robots.

Division of labor

A neural network is typically organized into layers, and each layer contains a large number of processing nodes. Data come in and are divided up among the nodes in the bottom layer. Each node manipulates the data it receives and passes the results on to nodes in the next layer, which manipulate the data they receive and pass on the results, and so on. The output of the final layer yields the solution to some computational problem.

In a convolutional neural net, many nodes in each layer process the same data in different ways. The networks can thus swell to enormous proportions. Although they outperform more conventional algorithms on many visual-processing tasks, they require much greater computational resources.

The particular manipulations performed by each node in a neural net are the result of a training process, in which the network tries to find correlations between raw data and labels applied to it by human annotators. With a chip like the one developed by the MIT researchers, a trained network could simply be exported to a mobile device.

This application imposes design constraints on the researchers. On one hand, the way to lower the chip’s power consumption and increase its efficiency is to make each processing unit as simple as possible; on the other hand, the chip has to be flexible enough to implement different types of networks tailored to different tasks.

Sze and her colleagues — Yu-Hsin Chen, a graduate student in electrical engineering and computer science and first author on the conference paper; Joel Emer, a professor of the practice in MIT’s Department of Electrical Engineering and Computer Science, and a senior distinguished research scientist at the chip manufacturer NVidia, and, with Sze, one of the project’s two principal investigators; and Tushar Krishna, who was a postdoc with the Singapore-MIT Alliance for Research and Technology when the work was done and is now an assistant professor of computer and electrical engineering at Georgia Tech — settled on a chip with 168 cores, roughly as many as a mobile GPU has.

Act locally

The key to Eyeriss’s efficiency is to minimize the frequency with which cores need to exchange data with distant memory banks, an operation that consumes a good deal of time and energy. Whereas many of the cores in a GPU share a single, large memory bank, each of the Eyeriss cores has its own memory. Moreover, the chip has a circuit that compresses data before sending it to individual cores.

Each core is also able to communicate directly with its immediate neighbors, so that if they need to share data, they don’t have to route it through main memory. This is essential in a convolutional neural network, in which so many nodes are processing the same data.

The final key to the chip’s efficiency is special-purpose circuitry that allocates tasks across cores. In its local memory, a core needs to store not only the data manipulated by the nodes it’s simulating but data describing the nodes themselves. The allocation circuit can be reconfigured for different types of networks, automatically distributing both types of data across cores in a way that maximizes the amount of work that each of them can do before fetching more data from main memory.

At the conference, the MIT researchers used Eyeriss to implement a neural network that performs an image-recognition task, the first time that a state-of-the-art neural network has been demonstrated on a custom chip.

READ MORE: http://news.mit.edu/2016/neural-chip-artificial-intelligence-mobile-devices-0203

ADDITIONAL: http://web.mit.edu/

The Trivium and the Quadrivium

The Liberal Arts

The liberal arts consists of seven branches of knowledge that prepare students for lifelong learning.

The concept is Classical, but the term Liberal Arts and their division into the TRIVIUM and the QUADRIVIUM date back to the Middle Ages.

The Trivium and the Quadriviumtrivium-wikipedia

The trivium includes those aspects of the liberal arts that pertain to mind.  The quadrivium, those aspects of the liberal arts that pertain to matter.

TRIVIUM= Logic, Grammar, and Rhetoric.

QUADRIVIUM= Arithmetic, Music, Geometry, and Astronomy.

Logic is the art of thinking; grammar, the art of inventing symbols and combining them to express thought; and rhetoric, the art of communicating thought from one mind to another, the adaptation of language to circumstance. Arithmetic, the theory of number, and music, an application of the theory of number (the measurement of discrete quantities in motion), are the arts of discrete quantity or number. Geometry, the theory of space, and astronomy, an application of the theory of space, are the arts of continuous quantity or extension.

liberal-arts-fig1

 

These arts of reading, writing, and reckoning have formed the traditional basis of liberal education, each constituting both a field of knowledge and the technique to acquire that knowledge. The degree bachelor of arts is awarded to those who demonstrate the requisite proficiency in these arts, and the degree master of arts, to those who have demonstrated a greater proficiency.

Today, as in centuries past, a mastery of the liberal arts is widely recognized as the best preparation for work in professional schools, such as those of medicine, law, engineering, or theology.

Those who first perfect their own faculties through liberal education are thereby better prepared to serve others in a professional or other capacity.

 

 

Source: The Trivium: The Liberal Arts of Logic, Grammar, and Rhetoric
by Sister Miriam Joseph Rauh, C.S.C., (1898-1982) earned her doctorate from Columbia University.
A member of the Sisters of the Holy Cross, Sister Miriam was professor of English at Saint Mary’s College from 1931 to 1960.

 

thinker_artist-rodin

During the era of classical antiquity (when ancient Greece and ancient Rome intertwined creating the Greco-Roman world), liberal arts was considered essential education for a free individual active in civic life. At the time, this would have entailed being able to participate in public debate, defend oneself and serve in court and on juries, and perform military service. At this time, liberal arts covered only three subjects: grammar, rhetoric and logic, collectively known as the trivium. This was extended in medieval times to include four further subjects: arithmetic, geometry, music and astronomy, named the quadrivium – so there were seven liberal arts subjects in the medieval liberal arts curriculum.

The trivium was considered preparatory work for the considerably more difficult quadrivium, with the quadrivium in turn being considered preparatory work for the more serious study of philosophy and theology. The aim of a liberal arts education was to produce a person who was virtuous and ethical, knowledgeable in many fields and highly articulate.

Although modern liberal arts curriculums have an updated choice of a larger range of subjects, it still retains the core aims of the liberal arts curricula maintained by the medieval universities: to develop well-rounded individuals with general knowledge of a wide range of subjects and with mastery of a range of transferable skills.

They will become ‘global citizens’, with the capacity to pursue lifelong learning and become valuable members of their communities.

Original Source: http://www.topuniversities.com/blog/what-liberal-arts-education

The Great Lesson in Liberal Arts

The Thinker- Liberal Arts Education

How does a liberal arts education help develop an individual’s character? As a working class individual with degrees in the humanities, I have argued the notion that my area of study helped me become a better person rather than a person with hard skills.

I have skills in writing, analysis, logical thinking, and the big one: empathy. I have come to terms that a liberal arts education is not a stepping-stone in a fully ordered career plan. However, I do have imagination and imagination is what I found to be the common trait in students that have studied Liberal Arts.

1) What makes a Liberal Arts Student Different

These individuals­ appeal to a certain type of imagination: empathy. For what is empathy if not imagination?

You imagine yourself in someone else’s shoes, you imagine “what if that starving child was my own child?” you imagine yourself closer to the atrocities being done even when your fifteen million miles away. Empathy is a form of imagination that demands you to stand outside of yourself and connect with another being.

2) The Lesson in Liberal Arts

Studying the humanities communicates the same message. Are we responsible to individuals we don’t even know, that are half way across the world?

The answer is yes. My duty as a moral individual is to help anyone in need, to be my brother’s keeper, to empathize with those who are suffering. My background in history and philosophy taught me not only that history repeats itself, but that you can learn from it too.

3) The Benefits of Liberal Arts

History may be tales of the victors, but future generations can learn from past mistakes. I began to realize that violence or self-shame are not choices that will aid me in my life journey.

I learned from my studies that life is unfair, yet there is hope because there are people who can empathize, imagine, and act. I have learned that value must be based on its moral implications and its ability to develop an individual into a better citizen of the world.

4) Why learning about the Liberal Arts is important

Why is learning of and about the Liberal Arts important? Because in this time and age, we need to teach our generation and the next one that a person’s value is not measured by utility. Teaching and learning the Liberal Arts will not fix everything, it will not cure or nullify the tragedies in the world.

However, it can provide a demarcation towards understanding. An understanding that will ultimately lead to empathy and action.

Given our current state in the world, from terrorism to environmental challenges, the ability to recognize “the other”- the “other” less fortunate, the “other struggling”- catapults an individual towards passionate change.

Knowing this, I highly recommend that anyone that reads this article share it on social media. Not just to learn about Liberal Art studies, but to discover the benefits of learning about the depths of human history.

Source: http://www.saycampuslife.com/2016/02/03/the-great-lesson-in-liberal-arts-what-we-can-all-takeaway/

New Learning Times Profile

New Learning Times NLT_LOGOThank you to George Nantwi and New Learning Times for their featured profile!

KevinCorbett_NewLearningTimesProfileIn the interview, I was asked five questions:

  1. How did your educational trajectory and past professional experience shape your current work?
  2. How do you hope your work will change the learning landscape?
  3. What broad trends do you think will have the most impact on learning in the years ahead?
  4. What are you currently working on & what is your next big project?
  5. Who are the most interesting people you are following on Twitter?

READ the entire Profile interview at New Learning Times: https://newlearningtimes.com/cms/article/3071/kevin-corbett

Or, download now the PDF: New Learning Times Profile: Kevin Corbett

About New Learning Times

The New Learning Times (NLT) provides daily coverage of the transformation of learning opportunities in the information age for those shaping the future of education. NLT is produced at the EdLab at Teachers College, Columbia University.

The editorial frame for NLT is governed by our understanding of three major trends, which we have termed “The New 3R’s.” Far beyond mere reform, the education sector is undergoing a major Reformation, a profound reconfiguration of the customs, institutions, and relationships that together constitute the foundations for learning opportunities around the world. Spurred by rapid developments in communications and computation, the education sector is also experiencing a Renaissance of new ideas, processes, and possibilities to support learning across the lifespan. The rapid introduction and convergence of these emerging political, technical, and artistic forces is creating the conditions for a Revolution in what is becoming the new learning sector. The New Learning Times seeks to chronicle the major transformation in learning possibilities.

Your Front Seat for the Education Revolution

New Learning Times is a mobile publication about today’s learning landscape covering the latest innovations in education and learning. Content includes edtech reviews, profiles of learning luminaries, and coverage of new ways to contribute to a learning revolution.

Flipped Classroom (#Infographic)

The definition of BLENDED LEARNING is a formal education program in which a student learns at least in part through online-learning, with some element of student control over time, place, path, and/or pace; at least in part in a supervised brick-and-mortar location away from home; and the modalities along each student’s learning path within a course or subject are connected to provide an integrated learning experience.

The Flipped Classroom is one approach to Blended Learning

The following Infographic on The Flipped Classroom is from Knewton (click for full-size)

flipped-classroom INFOGRAPHIC

 

More on Blended Learning & Flipping the Classroom:

About Knewton: https://www.knewton.com/

Knewton.com is the world’s most powerful adaptive learning engine. Knewton.com figures out what each student knows and how each student learns best, to pinpoint the type of content, level of difficulty, and which media format each student needs. Its technology can take any free open content, algorithmically calibrate it, and bundle it into a uniquely personalized lesson for each student at any moment.