Happy New Year!

Happy New Year! How was your winter break so far? Have you got some fun and got the chance to look into our first project? I just finished my first phase of study in Artificial Neural Network (ANN), and here is my note in PowerPoint form. You can use it to help you understand the basics. My next phase is to work on the Perceptron. If you had made some progress, feel free to share with us by posting your results in the blog. We will find time to meet as a group in the new year.

Ted Talk (implications of computers that can learn):

 Arthur Samuel was the “father” of machine learning because in 1956 he wanted to get a computer to beat him at checkers. He had computer play against itself and learn through thousands of times. In the end it worked.

 There is such a thing as a machine learning practitioner. Google shows it is possible to find information using a computer algorithm. This algorithm uses machine learning. Amazon, netflix use machine learning to suggest products. Rather than being programed by hand, algorithms learned how to do this from data.

 Self driving cars are possible because of machine learning. The self driving car has driven a million miles without any accident. Deep learning is an algorithm inspired by how the human brain works. Theoretically, there are no limitations to what it can do. The more data and computation time it has, the better it gets.

 Computers can listen and understand. They can also see. There is a 6 percent error rate in image recognition in computers. More complex sentences are understandable because of deep learning. Nuanced sentences are understandable. Computers can recognize what sentences are about. Computers can also write. They can describe images without having seen them in that order before. Deep learning can cut a lot of time from tasks.

Machine Learning Project 1: Perceptron

In our first task, we are going to learn and implement a basic neural network - "perceptron". In machine learning, the perceptron is an algorithm for supervised learning. It is a type of linear classifier which can classify linearly separable data. The following links give you a starting point to study perceptron. You are more than welcome to share any resource you found online with us.

Tutorials:

• How the Perceptron Algorithm Works 1/2  
• How the Perceptron Algorithm Works 2/2
• Java Implementation of the Perceptron Algorithm
• Understanding Multi-Layer Perceptron (MLP) .. How it Works
• Artificial Neural Networks (Part 1) - Classification using Single Layer Perceptron Model
• Artificial Neural Networks (Part 2) - -Classification using Multi-Layer Perceptron Model
• Artificial Neural Networks (Part 3) - Backpropagation
• Perceptrons for Dummies

Goals:

• Understand the basic operation of Perceptron
• Backpropagation
• Learning rules of Perceptron
• Limitations of Perceptron
• Implement a Single-Layer Perceptron (SLP)
• Real-world applications of Perceptron
• Implement a Multi-Layer Perceptron (MLP)

Target Dates:

• Study Phase: 12/10/2015 - 12/23/2015
• Implementation Phase: 12/24/2015 - 01/03/2015
• Seminar: 01/08/2015

Deep Learning and Big Data

Machine Learning is no longer a fancy si-fi dream or an obscure research field in the dark corner of the ivory tower. It is current a highly competitive research field for both academia and industry. Industry giants such as Google, Facebook, and Baidu have acquired most of the dominant players in this space to improve their services. Small start-ups also try eagerly to find a niche in this new domain. Watch a video of a panel discussion about "Deep Learning: Intelligence from Big Data". hosted by a talented speaker venture capitalist Steve Jurvetson. The discussion was located at Stanford Graduate School of Business on September 16, 2014.

Welcom to Machine Learning! ..... Schedule Our Kick-off Meeting

Welcome to the Machine Learning Research Group! So far, we have 11 students (both sophomores & juniors) joining the group. The goals of the group are to tap into the cutting-edge academic & commercial resources of Artificial Intelligence (AI), follow up the development & trends of the new technologies, acquire the theoretical background and practical skills of machine learning, and apply machine learning algorithms to solve real-world problems.

Most of the early-stage research activities will be focusing on the learning curve of this new technology. It will include 

• basic concepts (definition, background, scope, structure, methods & applications)
• neural networks and learning algorithms,
• programming languages & machine learning libraries, and
• brainstorming the potential applications.

The second-stage research activities will be focusing on a few challenging research projects . We will divide our group into a few teams to tackle each challenge. These projects may lead to competitions (such as Google Science Fair), poster presentations in professional conferences (such as IEEE ISEC), publishing of research papers in research journals (such as NHSJS), class projects in Advanced STEM Research (see current STEM course blog), filing patents or commercial applications. Only sky will be the limit!

Most of our "meetings" will be online through the blogs or any electronic media. That's why I have invited all of the group members to be the authors of the Machine Learning blog (http://stem-ml.blogspot.com/). You should contribute to the blog frequently by posting and commenting on new development, video links, programming tutorials, study notes, and research results. You should also visit the blog frequently for information, activities or assignments. The blog will be our main playground for collaboration. If you haven't received the invitation, please let me know as soon as possible. We will also arrange some "meet-up time" at school at everyone's convenience. For our first "kick-off meeting", please use the link to indicate your available time slots. Please response by 8:00  PM (12/06/2015) tonight. If it's hard to find common meeting time for everyone, we will try to meet in subgroups.

In order for our research group to be productive, we will create a few assignments and set some milestones to help everyone pick up the new skills. The first assignment will be about a simple neural network called "Perceptron". The details will be coming soon. In the mean time, please enjoy the videos I have posted in our blog, and share your thoughts or post your notes. These videos will provide you the basic concepts of machine learning. If you come across any good video, please share with us. I hope to meet you all in our kick-off meeting! 

Happy Researching!

What Neural Network Can Do for You?

There are growing number of applications in the machine learning field. Here is one of the video: writes sentence about images. While you are learning more and more about the machine learning applications, you can start thinking  about your own applications. What can neural network do for you? What can machine learning change certain aspects of your life? Share with us your ideas!