MIST101 Workshop #1 Event Review

Author: Alex Li,  IEEE

Reviewer: Justin Yuan, UTMIST & IEEE


Pasted image at 2017_09_20 04_01 PM

This was the first machine learning workshop put on by UTMIST in association with IEEE. As the first of six workshops, it served as an introduction to many of the topics that will be covered in more detail in future workshops. This post will give an overview of the topics covered in this workshop.

 

IMG_9645

There are traditionally two approaches to problem-solving: following a set of instructions or learning from experience. Traditionally, a computer program is in the former category. It is a set of instructions, a recipe tailor-made for a single problem that needs to be created by a human expert. For some problems, however, it is impossible or extremely difficult to cover all potential circumstances by hand. Machine learning takes the second approach, to learn from experience. By designing an algorithm that can learn from examples and solve a family of tasks, we can solve a wide variety of problems that were previously inaccessible. The many applications of machine learning include classification, pattern recognition, recommender systems, self-driving cars, natural language processing, computer vision, robotics, and playing games better than any human being ever could.

 

 

 

 

 

In general, there are three types of machine learning tasks: supervised learning, unsupervised learning, and reinforcement learning.

IMG_9626Supervised learning attempts to learn the underlying function of given input/output pairs. These types of functions can take a variety of forms. For example, an image classifier maps pixels to object categories, and a translator maps one language to another. Machine learning tries to create models to approximate these functions. These models include decision trees, graphical models, Gaussian processes, SVMs, and KNNs. Recently, artificial neural networks have supplanted many of these methods and will therefore be a major topic of discussion in the workshops. The different types of artificial neural networks include feedforward neural networks, convolutional neural networks, and recurrent neural networks. More specifics will be given in later workshops.

 

 

 

 

IMG_9615

 

In general, supervised learning is marked by the presence of already known data-label pairs, which are fed into algorithms that enables effective model learning. Unsupervised learning, on the other hand, does not have any known labels to begin with. Some common applications of unsupervised learning are clustering, which can, for example, group images together that are similar, semi-supervised learning, which applies concepts from both supervised and unsupervised learning, and dimensionality reduction, which can reduce the number of variables
under consideration in a problem.

 

 

 

 

 

IMG_9646

 

 

Finally, there is reinforcement learning, which is used for a very different type of situation. It essentially involves an agent playing a game and improving its performance by reinforcing better behaviors. This type of learning doesn’t require any correct or ideal cases to be given to it, but rather only some type of judgment as to whether it is performing well. It works by exploring a set of actions randomly or under proper guidance, evaluating them as to how well they perform in terms of a “reward”, adjusting its set of actions, and then repeating this entire process. Reinforcement learning requires little human expertise but can perform extremely intelligent behaviors. However, it requires extensive training and isn’t very adaptable to new situations.

 

 

 

 

 

 

In future workshops, all of these topics will be covered in a more in-depth and rigorous manner. We hope you come and attend them! The next workshop will be on supervised learning and neural networks and will occur in the week of Oct. 2-6.

 

You may find the slides for the workshop here!

MIST101 #1: Introduction to Machine Intelligence

Want to learn more about machine learning, and don’t know where to start? We have got you! Come join us at our MIST101, the ultimate all-in-one package of introductory workshops to machine learning! Come to our first workshop on Monday, Sept 18, 6-7 pm OR 7-8 pm at BA 2159! Two identical sessions (i.e. 6-7pm and 7-8pm) will be hosted to accommodate more audience, please only sign up for one session!

Space is limited and will be on a first-come first-serve basis. Please RSVP by depositing $5 through eventbrite. The full ticket price will be refunded after the event for all attendees who showed up.

Thank you for your support and looking forward to meeting you at the event!

UTMIST Banner Photo

Introducing UTMIST!

  1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.

  2. A robot must obey orders given it by human beings except where such orders would conflict with the First Law.

  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

                                                                                I, Robot, by Issac Asimov


In the past we used to think that artificial intelligence only existed within scientific fictions, in the wild imagination and fantasy of their authors.

However, the success of Alpha Go has completely upended our thinking in this regard and showed us: Artificial Intelligence has permeated our daily lives!

How much do you really know about artificial intelligence, though?

Let UTMIST tell you about it all!


We are UTMIST

The University of Toronto, as one of the top education institutes in the world, leads the way in the field of artificial intelligence, and it has Geoffrey Hinton’s Machine Learning Group – the first of the three world-leading forces in the field of deep learning. However, undergraduate students seldom have the opportunity of getting in touch with the scholars from the Machine Learning Group, or getting exposure to leading technology in machine learning. This has created a disconnection between the undergraduate students and top scholars in machine learning. UTMIST aims to clear the mist by demystifying artificial intelligence technology, which is the most trendy technology nowadays. As a student organization focused on machine learning research, UTMIST is to establish a close connection between the undergraduate students and top academic resources at U of T.

University of Toronto Machine Intelligence Student Team (UTMIST) is an officially certified student organization within the University of Toronto. Our mission is to let more people get to know about artificial intelligence, “Clear the mist”!

 


In order to help everyone learn machine learning in a systematic way, UTMIST here presents our carefully prepared workshop series – “MIST101”

MIST101

During the initial stages, UTMIST will host a series of workshops on machine learning and data science – “MIST101”, and show everyone the most advanced, popular and upfront technology.

The workshop will be hosted on a bi-weekly basis, and the content will be prepared by our outstanding academic team. The workshop will cover supervised learning, unsupervised learning, TensorFlow and more. We will start from the basic theories and extend to the most cutting-edge research. Our instructors will lead everyone from the basics into the deeper understanding of what machine learning is and how it is to be applied in real life. You will get the opportunity to interact with our instructors, getting face-to-face guidance, and there may even be an opportunity for you to join our instructor’s research project!

UTMIST welcomes all undergraduate students who are interested in machine learning. Even if you are not planning to pursue a career in machine learning, everything you learn from MIST101 will still give you a head start in interacting with the field in artificial intelligence!


MIST Academy

Sheldon-cropped

Sheldon Huang

  • UTMIST President and co-founder.
  • Recent Research in the Machine Learning Group at U of T: Generative Adversarial Network, Optimization, Style and Domain Transfer.
  • TA in the Mathematics Department at University of Toronto. He was a calculus TA in his second year, and a linear algebra TA in his third year.
  • Started his research career in Grade 9. Some of his other researches include: real analysis, non-linear optics and life sciences. Published his first paper during the summer after Grade 10.
  • Currently a 3rd year Computer Science student at University of Toronto.  

Justin-cropped

Justin Yuan

  • UTMIST co-founder and project director.
  • Third year undergraduate student; Bachelor of Engineering Science Candidate (Robotics Major).
  • 2017 Cansat Competition U of T Team, program team lead.
  • Research intern at the D3M lab; focuses include image classification, text classification and attention mechanism in deep learning.
  • Current IEEE UofT student branch, event director of computer chapter.
  • University of Toronto application development association senior developer.
  • Technical analyst at the renowned media company Synced Review (global).

Colin-cropped

Colin Li

  • UTMIST co-founder and project director.
  • Third year undergraduate student; Bachelor of Engineering Science Candidate (Robotics Major).
  • CSC373 (Algorithm Design, Analysis and Complexity) TA.
  • 2017 ACM NA regional competition and 2017 NAIPC competition top 10
  • Participated in the research project of using deep learning techniques to improve the trajectory tracking performance of quadrotors.
  • Published research paper at IEEE international robotics and automation conference.
  • Lead author of the published paper – “ Deep Neural Network for Improved, Impromptu Trajectory Tracking of Quadrotors” – Q. Li, J.Qian, Z.Zhu, X.Bao, M.K.Helwa and A.P. Schoellig.

xuechen-Facebook

Xuechen Li

  • UTMIST project director.
  • Third year undergraduate student major in computer science, mathematics and statistics.
  • Summer research project: Bayesian neural network and variational inference.

Talk series

If MIST101 leaves you curious for more knowledge in machine learning, then our talk series will be your next step!

UTMIST will host research talks regularly, and we will have the most influential scholars as speakers to share their research results. The guest speakers to be invited will include graduate students, current technology company employees and professors. They will explain and analyze the most cutting-edge technology of machine learning from their own perspectives, and share their thoughts about the future trends in the field. Approaching the subject from various perspectives, our talk series will be able to help the attendants have a grasp of the latest development in machine learning in the most comprehensive way.

Most importantly, all our talk series are free and open to the public!


With our mission in mind, UTMIST appreciates all your support and aspires to present you with the best and most exciting contents!