What is Machine Learning?
How can a computer learn to diagnose cancer?
How can a robotic assistant learn to adapt to the specific habits of their owners?
Machine learning is the study of how computers can learn complex concepts from data and experience, and seeks to answer the fundamental research questions underpinning the challenges outlined above.
The field of machine learning crosses a wide variety of disciplines that use data to find patterns in the ways both living systems, such as the human body and artificial systems, such as robots, are constructed and perform.
Whether it’s being applied to analyze and learn from medical data, or to model financial markets, or to create autonomous vehicles, machine learning builds and learns from both algorithm and theory to understand the world around us and create the tools we need and want.
Specific Machine Learning research topics in Computer Science include learning conceptual structures through developmental processes; improving control of stochastic and nonlinear dynamic systems through reinforcement feedback; learning robot control strategies; finding patterns in large bodies of data represented in graphical form, including social networks; extracting or retrieving information in natural language; classification of genetic data; and using learning methods for improving discrete optimization algorithms.
Machine Learning is multi-disciplinary
Much of the machine learning research in Computer Science is multi-disciplinary, with strong ties to research in statistics, operations research, cognitive and developmental psychology, neuroscience, and philosophy.
To give you a better sense of courses offered under Machine Learning, let’s have a look at various courses offered by different departments at Duke University for Machine Learning.
#1 Computer Science Department
- Algorithmic Aspects of Machine Learning
- Computational Systems Biology
- Computer Vision
- Introduction to Artificial Intelligence
- Machine Learning
#2 Electrical and Computer Engineering
- Acoustics and Hearing
- Adaptive Filters
- Digital Image and Multidimensional Processing
- Digital Processing of Speech Signals
- Digital Signal Processing
- Fundamentals of Digital Signal Processing
- Information Theory
- Introduction to Digital Communication Systems
- Introduction to Robotics and Automation
- Introduction to Signals and Systems
- Linear Control Systems
- Random Signals and Noise
- Sensor Array Signal Processing
- Sound in the Sea: Introduction to Marine Bioacoustics
#3 Mathematics Department
- Applied Stochastic Processes
- Scientific Computing
- Stochastic Calculus
#4 Statistical Sciences Department
- Applied Stochastic Processes
- Computational Data Analysis
- Introduction to Statistical Methods
- Modeling and Scientific Computing
- Modern Nonparametric Theory and Methods
- Probability and Statistical Models
- Statistical Case Studies
For different schools, Machine Learning curriculum could be very different. So you should be close attention to Machine Learning curriculum.
The curriculum actually tells you what subjects you’ll be studying and straight away gives an impression about the relevance of the program for you.
How to Build Your Profile for MS in Machine Learning?
Thinking for pursuing an MS in Machine Learning (or, Data Science)?
Head to the Home of Data Science and Machine Learning – Kaggle Competition!
Kaggle is a platform for predictive modelling and analytics competitions in which companies and researchers post data and statisticians and data miners compete to produce the best models for predicting and describing the data. This crowdsourcing approach relies on the fact that there are countless strategies that can be applied to any predictive modelling task and it is impossible to know at the outset which technique or analyst will be most effective.
Kagglers come from a wide variety of backgrounds, including fields such as computer science, computer vision, biology, medicine, and even glaciology. It also includes many of the world’s best-known researchers, including members of IBM Watson’s Jeopardy-winning team and the team working on Google’s DeepMind. Many of these researchers publish papers in peer-reviewed journals based on their performance in Kaggle competitions.
How does Kaggle Competitions Works?
- Companies and organizations prepares the data and a description of the problem. Kaggle frame the competition, anonymize the data, and integrate the winning model into their operations.
- Participants, like you, experiment with different techniques and compete against each other to produce the best models. Work is shared publicly through Kaggle Scripts to achieve a better benchmark and to inspire new ideas. Submissions are made through Scripts or through private manual upload. For most competitions, submissions are scored immediately (based on their predictive accuracy relative to a hidden solution file) and summarized on a live leaderboard.
- After the deadline passes, the host company pays the prize money for the winning solution. many companies recruit participants based on their place on the leaderboard, final score, and submitted scripts.
- Alongside its public competitions, Kaggle also offers private competitions limited to Kaggle’s top participants.
What Kaggle competition should a beginner start with?
I’d start with the tutorials first just to make sure you have a good grasp of the primary tools and techniques that most people use: https://www.kaggle.com/wiki/Home
Afterwards, Titanic: Machine Learning from Disaster is a good competition to start. It will prep you with fundamentals of data science – the data size is manageable, the problem is interesting, and you need minimum overhead in terms of computational requirements.
If you aren’t decided on your weapon of choice, I would suggest that you start with R. The tutorial can be found at Titanic: Machine Learning from Disaster. Follow this up with Python, Titanic: Machine Learning from Disaster.
Since your objective is learning, the most important place for you is the Kaggle forum. There is just tons of valuable information buried in those posts. What worked, what didn’t work, the issues others are facing, interesting patterns and visualizations, and neat tricks. I find it to be the best “practical” data science guide out there.
Once you have a sound footing, maybe in a couple of weeks, the next step would be to try something with text data like Sentiment Analysis on Movie Reviews.
Add to that some competition that uses audio and/or video data. There could be a few running or you can always dig up the old ones like Challenges in Representation Learning: Facial Expression Recognition Challenge and The Marinexplore and Cornell University Whale Detection Challenge
Selecting the right school for Machine Learning
Let’s first have a look some of the good schools, in no particular order, offering Master’s in Machine Learning:
- Carnegie Mellon University
- University of Michigan Ann Arbor
- Columbia University
- University of Washington
- Georgia Tech
- University of California San Diego
- University of Massachusetts Amherst
- John Hopkins University
- University of Illinois Urbana Champaign
- Penn State University
- University of North Carolina Chapel Hill
- California Institute of Technology
- University of Wisconsin Madison
Now the question is, how to go about selecting the right school for you. Let me dive deep into some of the schools:
|Avg GRE Quant score||164|
|Ranking||#9 – US News Ranking for Artificial Intelligence – Computer Science|
|Description||The Center for Machine Learning at Georgia Tech (ML@GT) is an Interdisciplinary Research Center that is both a home for thought leaders and a training ground for the next generation of pioneers.|
The field of machine learning crosses a wide variety of disciplines that use data to find patterns in the ways both living systems, such as the human body and artificial systems, such as robots, are constructed and perform. Whether it’s being applied to analyze and learn from medical data, or to model financial markets, or to create autonomous vehicles, machine learning builds and learns from both algorithm and theory to understand the world around us and create the tools we need and want.
|Shortlist type||For a profile of 3.7+ GPA, GRE of 330+ with Quant score of 164, Decent research experience with 1 publication and 1 internship / project, I would shortlist this school as a Dream school i.e. with 25% chance of getting an admission.|
|Avg GRE Quant score||167|
|Ranking||#15 – US News Ranking for Artificial Intelligence – Computer Science|
|Description||The Machine Learning Track is intended for students who wish to develop their knowledge of machine learning techniques and applications. Machine learning is a rapidly expanding field with many applications in diverse areas such as bioinformatics, fraud detection, intelligent systems, perception, finance, information retrieval, and other areas.|
|Shortlist type||For a profile of 3.7+ GPA, GRE of 330+ with Quant score of 164, Decent research experience with 1 publication and 1 internship / project, I would shortlist this school as a Reach school i.e. with 50% chance of getting an admission.|
University of North Carolina Chapel Hill
|Avg GRE Quant score||157|
|Ranking||#25 – US News Ranking for Artificial Intelligence – Computer Science|
|Description||The problems we study combine vast amounts and disparate types of measurements with equally complex prior knowledge, posing unique challenges for machine learning. Our interests include both modeling paradigms, such as Bayesian nonparametric methods, and inference methodologies, such as MCMC, variational methods and convex optimization.|
|Shortlist type||For a profile of 3.7+ GPA, GRE of 330+ with Quant score of 164, Decent research experience with 1 publication and 1 internship / project, I would shortlist this school as a Safe school i.e. with 75% chance of getting an admission.|
I recommend finding 4-6 Dream, Reach and Safe schools as above.
Once you figure out your schools, you would research a bit about professors and alumni in those schools. It’s your individual fit with an advisor that matters more than anything else.
Some great professors in Machine Learning are at less known schools. So, starting with the the list of schools listed above, check-out the professors and their research areas that interest you most. You may look over some of their papers or try to talk to their students to get a feeling for the intellectual and inter-personal style in their group. You may also contact them by e-mail to discuss your research interests.
In addition, you should clearly state your research interests in the statement of purpose and note your interest in being part of the Machine Learning at your target school.
Career in Machine Learning
While you are thinking about pursuing your MS in Machine Learning, it’s important to get a sense of career in Machine Learning.
As Machine Learning is still evolving, after a Master’s in Machine Learning, you are more likely to, and are better off, work in some research areas in Machine Learning.
You are expected to solve new and emerging technical challenges related to human-machine interactions.
In your role, you will utilize core computer science and engineering skills like high-performance computing, distributed systems and applied math.
You are expected to have 5+ years of experience in programming parallel and distributed systems, debugging low-level problems, performance analysis and optimizations, and numerical methods.
Also include – experience in using machine learning techniques for classification, regression, or ranking problems, experience in building predictive models for recommendations or personalization, design and implementation of shipping, innovative consumer products etc.
Typical employers include Facebook, Amazon, Apple, Google and Microsoft.
Hope this helps in shortlisting your target schools.
If you need more hand-holding and guidance, I have a solution for you. Check-out our Premium Counseling, in which we would customize shortlisting for your profile and interests. We would also work closely with you in writing your essays, submitting your applications and preparing for visa interviews.