MIT: Independent Activities Period: IAP

IAP 2018 Activities by Category - Mathematics

= Add activity session to your calendar (exports in iCalendar format)
Expand All | Collapse All


Causal Inference & Deep Learning

Max Shen, PhD Student, Computational & Systems Biology

Enrollment: Unlimited: No advance sign-up
Attendance: Participants welcome at individual sessions

The vast majority of machine learning advances concern associative or correlative relationships despite the importance of learning and inferring causal relationships. 

In this brief class, we will consider how the successful tool of deep learning can best deliver new insights into fundamental problems regarding causality. We will explore several recent and successful papers describing applications of deep learning to causality, with the big picture goal of understanding fruitful next steps in the research intersection of deep learning and causality.

Please register here for class email updates: https://goo.gl/forms/xOgpVJB5OFsrUB6G3

Contact: Max Shen, MAXWSHEN@MIT.EDU


Death to Riemann! Long Live Minkowski using Quaternions!

Douglas Sweetser '84

Add to Calendar Jan/25 Thu 03:00PM-05:00PM 3-333

Enrollment: Unlimited: Advance sign-up required

Einstein asked his great buddy and math savant Marcel Grossman for a flexible math tool for geometry. Marcel went to the library and returned the next day with his answer: Riemann geometry. It was a great answer even if arcain and has ruled the road ever since. Yet not efforts to unify gravity with the rest of physics have worked.

In this two hour jam session, we challenge Grassman's suggestion by working with a number that can be added, subtracted, multiplied, or divided and has 4 part harmony, the quaternions. Leonard Susskind's three books from The Theoretical Minimum series will be our guide. Always having four parts to every written expression no matter how simple or complex is odd but opens many new views on mathematical physics. Register today!

Sponsor(s): Alumni Association
Contact: Elena Byrne, W98-206C, 617 252-1143, EBYRNE@MIT.EDU


Mathematics Lecture Series

Alan Edelman

Enrollment: Unlimited: No advance sign-up
Attendance:

Ten lectures by Mathematics faculty members on interesting topics from both classical and modern Mathematics. All lectures should be accessible to students with a Calculus background and an interest in Mathematics. These lectures are open to the public and you may attend as many or as few as you wish.  Students wishing to receive course credit for attending the full set of 10 lectures should consider registering for the IAP for-credit subject 18.095, in which P-sets are assigned and students prepare these for discussion in a weekly problem session.

For more information on idividual lectures please see:

http://math.mit.edu/academics/iap.php

Sponsor(s): Mathematics
Contact: Alan Edelman, edelman@mit.edu


Add to Calendar Jan/08 Mon 01:00PM-02:30PM 2-190, Speaker: Heather Macbeth
Add to Calendar Jan/10 Wed 01:00PM-02:30PM 2-190, Speaker: Gil Strang
Add to Calendar Jan/12 Fri 01:00PM-02:30PM 2-190, Speaker: Steven Johnson
Add to Calendar Jan/17 Wed 01:00PM-02:30PM 2-190, Speaker: Jeremy Kepner
Add to Calendar Jan/19 Fri 01:00PM-02:30PM 2-190, Speaker: Dan Stroock
Add to Calendar Jan/22 Mon 01:00PM-02:30PM 2-190, Speaker: Juan Pablo Vielma
Add to Calendar Jan/24 Wed 01:00PM-02:30PM 2-190, Speaker: Thomas Beck
Add to Calendar Jan/26 Fri 01:00PM-02:30PM 2-190, Speaker: Scott Sheffield
Add to Calendar Jan/29 Mon 01:00PM-02:30PM 2-190, Speaker: Philippe Rogollet
Add to Calendar Jan/31 Wed 01:00PM-02:30PM 2-190, Speaker: Joern Dunkel

For more information on individual lectures please see:

http://math.mit.edu/academics/iap.php

 

 


Mathematics of Big Data & Machine Learning

Jeremy Kepner, Fellow & Head MIT Supercomputing Center

Enrollment: Limited: Advance sign-up required
Sign-up by 12/22
Limited to 20 participants
Attendance: Participants must attend all sessions
Prereq: Linear Algebra

Big Data describes a new era in the digital age where the volume, velocity, and variety of data created across a wide range of fields (e.g., internet search, healthcare, finance, social media, defense, ...)  is increasing at a rate well beyond our ability to analyze the data.  Machine Learning has emerged as a powerful tool for transforming this data into usable information.  Many technologies (e.g., spreadsheets, databases, graphs, linear algebra, deep neural networks, ...) have been developed to address these challenges.  The common theme amongst these technologies is the need to store and operate on data as whole collections instead of as individual data elements.  This class describes the common mathematical foundation of these data collections (associative arrays) that apply across a wide range of applications and technologies.  Associative arrays unify and simplify Big Data and Machine Learning.  Understanding these mathematical foundations allows the student to see past the differences that lie on the surface of Big Data and Machine Learning applications and technologies and leverage their core mathematical similarities to solve the hardest Big Data and Machine Learning challenges.

 

Sponsor(s): Mathematics
Contact: Jeremy Kepner, 2nd Floor, 300 Tech Sq, 781 981-3108, KEPNER@LL.MIT.EDU


Manipulation Big Data

Add to Calendar Jan/09 Tue 11:00AM-01:00PM 300 Tech Sq Floor 2

Chapters 1 and 2 of "Mathematics of Big Data" text.

Jeremy Kepner - Fellow & Head MIT Supercomputing Center


D4M: A New Tool for Big Data

Add to Calendar Jan/16 Tue 11:00AM-01:00PM 300 Tech Sq Floor 2

Chapter 3 and Chapter 4 of "Mathematics of Big Data" text.

Introduction to D4M (http://d4m.mit.edu).

Jeremy Kepner - Fellow & Head MIT Supercomputing Center


Four Perspectives on Data

Add to Calendar Jan/23 Tue 11:00AM-01:00PM 300 Tech Sq Floor 2

Chapter 5 and 6 of "Mathematics of Big Data" text.

Jeremy Kepner - Fellow & Head MIT Supercomputing Center


Mathematical Foundations of Big Data

Add to Calendar Jan/30 Tue 11:00AM-01:00PM 300 Tech Sq Floor 2

Chapters 7 and 8 of "Mathematics of Big Data".

Jeremy Kepner - Fellow & Head MIT Supercomputing Center


MathWorks: Demystifying Deep Learning - A Practical Approach in MATLAB

JM.Modisette, PhD, Technical Evangelist

Add to Calendar Jan/24 Wed 10:00AM-12:00PM 3-270

Enrollment: Register on MathWorks Website (below)
Limited to 119 participants
Prereq: None

Are you new to deep learning and want to learn how to use it in your work?  Deep learning can achieve state-of-the-art accuracy in many humanlike tasks such as naming objects in a scene or recognizing optimal paths in an environment.

The main tasks are to assemble large data sets, create a neural network, to train, visualize, and evaluate different models, using specialized hardware - often requiring unique programming knowledge. These tasks are frequently even more challenging because of the complex theory behind them.

In this seminar, we’ll demonstrate new MATLAB features that simplify these tasks and eliminate the low-level programming. In doing so, we’ll decipher practical knowledge of the domain of deep learning.  We’ll build and train neural networks that recognize handwriting, classify food in a scene, and figure out the drivable area in a city environment.  

For more information and registration at:

https://www.mathworks.com/company/events/seminars/mit-iap-2361872.html

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


MathWorks: Introduction to MATLAB: Problem Solving and Programming

JM.Modisette, Phd, Technical Evangelist

Add to Calendar Jan/22 Mon 10:00AM-02:00PM 32-141, Attendees should bring a laptop

Enrollment: Register on MathWorks Website (below)
Limited to 90 participants
Prereq: None

In this hands-on workshop, you will learn how to import data from an external file, plot the data over time, then perform some analysis to view the data trends.  You’ll learn how to write a MATLAB script and publish it to a format for sharing, such as HTML. You’ll also learn how to write your own MATLAB functions, use flow control, and create loops.

By the end of the session, you’ll have learned to create an application in MATLAB.

 

Register at: https://www.mathworks.com/company/events/seminars/mit-iap-2341781.html 

Note: Attendees should bring a laptop to this hands-on lab.

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


MathWorks: Introduction to Simulink

JM.Modisette, Phd, Technical Evangelist

Add to Calendar Jan/22 Mon 02:30PM-06:30PM W31-301

Enrollment: Register on MathWorks Website (below)
Limited to 30 participants
Prereq: No Simulink experience is assumed or necessary.

This workshop is a 4-hour hands-on tutorial of Simulink, the block diagram environment integrated with MATLAB for multidomain simulation and design.

During the workshop, you will follow along with the presenter in creating a Simulink model from scratch and then building upon and improving that model throughout the session. You’ll get an introduction to many Simulink features, including:

- Modeling a dynamic system

- Logic operations

- Model hierarchy

- Masks

- Creating custom libraries

- Model referencing

- Vectorization

 

Learn how Simulink is used for the following applications:

- Modeling continuous systems

- Control system development

- Algorithm design and simulation

 

 Register at: https://www.mathworks.com/company/events/seminars/mit-iap-2341781.html%20.html

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


MathWorks: Parallel and Distributed Computing with MATLAB Seminar

JM.Modisette, Phd, Technical Evangelist

Add to Calendar Jan/23 Tue 10:00AM-12:00PM 3-270

Enrollment: Limited: First come, first served (no advance sign-up)
Limited to 119 participants
Prereq: None

Large-scale simulations and data processing tasks that support engineering and scientific activities such as mathematical modeling, algorithm development, and testing can take an unreasonably long time to complete or require a lot of computer memory. You can speed up these tasks by taking advantage of high-performance computing resources, such as multicore computers, GPUs, computer clusters, and cloud computing services.

Using the Parallel Computing capabilities in MATLAB allows you to take advantage of additional hardware resources that may be available either locally on your desktop or on clusters and clouds. By using more hardware, you can reduce the cycle time for your workflow and solve computationally and data-intensive problems faster.   

We will discuss and demonstrate how to perform parallel and distributed computing in MATLAB. We will introduce you to parallel processing constructs such as parallel for-loops, distributed arrays, and message-passing functions. We will also show you how to take advantage of common trends in computer hardware, from multiprocessor machines to computer clusters.

Highlights Include:

-Built-in support for parallel computing

-Creating parallel applications

-Scaling up to computer clusters, grid environments or clouds

-Employing GPUs

-Programming with tall and distributed arrays to work with large data sets

 

Register at: https://www.mathworks.com/company/events/seminars/mit-iap-2362013.html

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


MathWorks: Parallel and Distributed Computing with MATLAB Workshop

JM.Modisette, PhD, Technical Evangelist

Add to Calendar Jan/23 Tue 01:00PM-04:00PM W31-301

Enrollment: Register on MathWorks Website (below)
Limited to 30 participants

Following the Parallel and Distributed Computing Seminar, this workshop will allow attendees to practice different techniques for parallel and distributed computing.

Highlights include:

- 5-minute introduction to resources at MIT

- 20-minute overview of Parallel Computing with MATLAB

- 1.5 hour hands-on, self-paced workshop

- 30-minute Q&A session of workshop

- 30-minute hands-on session of submitting job to cluster

 

Register at: https://www.mathworks.com/company/events/seminars/mit-iap-2362013.html

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


MathWorks: Practical Applications of Deep Learning, A Hands-On MATLAB Workshop

JM.Modisette, PhD, Technical Evangelist, MathWorks

Add to Calendar Jan/24 Wed 01:00PM-04:00PM W31-301

Enrollment: Register on MathWorks Website (below)
Limited to 30 participants

Are you new to deep learning and want to learn how to apply these techniques it in your work? Deep learning achieves human-like accuracy for many tasks considered algorithmically unsolvable with traditional machine learning. It is frequently used to develop applications such as face recognition, automated driving, and image classification.

In this hands-on workshop, you will write code and use MATLAB to:

- Learn the fundamentals of deep learning and understand terms like “layers”, “networks”, and “loss”

- Build a deep network that can classify your own handwritten digits

- Access and explore various pretrained models

- Use transfer learning to build a network that classifies different types of food

- Train deep learning networks on GPUs in the cloud

- Learn how to use GPU code generation technology to accelerate inference performance

 

Register at: https://www.mathworks.com/company/events/seminars/mit-iap-2361872.html

Sponsor(s): Office of Educational Innovation and Technology
Contact: JM.Modisette, JM.Modisette@mathworks.com


More on algebraic topology

Sanath Devalapurkar

Enrollment: Unlimited: No advance sign-up
Attendance: Participants welcome at individual sessions
Prereq: 18.906/math 231b, spectral sequences

The goal of this seminar/course will be to describe certain topics in homotopy theory that extend the topics discussed, for instance, in the Kan seminar (18.919) and the spectral sequences seminar (which met earlier this fall). All updates (time, location, syllabus, etc) will be at http://www.mit.edu/~sanathd/chromotopy.html --- see there for more information!

Topics to be covered will include: (basic computations with) the Adams(-Novikov) spectral sequence, formal group laws and Quillen's theorem, the Ravenel conjectures, Morava K- and E-theories, and (if there's time) the construction of TMF.

All updates (time, location, syllabus, etc) will be at http://www.mit.edu/~sanathd/chromotopy.html --- see there for more information!

Contact: Sanath Devalapurkar, SANATHD@MIT.EDU


Algebraic topology seminar

Jan/08 Mon 02:00PM-03:00PM (CANCELED)
Jan/09 Tue 02:00PM-03:00PM (CANCELED)
Jan/10 Wed 02:00PM-03:00PM (CANCELED)
Jan/11 Thu 02:00PM-03:00PM (CANCELED)
Jan/12 Fri 02:00PM-03:00PM (CANCELED)
Add to Calendar Jan/16 Tue 02:00PM-03:00PM Location TBD
Add to Calendar Jan/17 Wed 02:00PM-03:00PM Location TBD
Add to Calendar Jan/18 Thu 02:00PM-03:00PM Location TBD
Add to Calendar Jan/19 Fri 02:00PM-03:00PM Location TBD
Add to Calendar Jan/22 Mon 02:00PM-03:00PM Location TBD
Add to Calendar Jan/23 Tue 02:00PM-03:00PM Location TBD
Add to Calendar Jan/24 Wed 02:00PM-03:00PM Location TBD
Add to Calendar Jan/25 Thu 02:00PM-03:00PM Location TBD
Add to Calendar Jan/26 Fri 02:00PM-03:00PM Location TBD
Add to Calendar Jan/29 Mon 02:00PM-03:00PM Location TBD
Add to Calendar Jan/30 Tue 02:00PM-03:00PM Location TBD
Add to Calendar Jan/31 Wed 02:00PM-03:00PM Location TBD
Add to Calendar Feb/01 Thu 02:00PM-03:00PM Location TBD
Add to Calendar Feb/02 Fri 02:00PM-03:00PM Location TBD

See http://www.mit.edu/~sanathd/chromotopy.html for updates!

Sanath Devalapurkar


ORC IAP Seminar 2018: "Operations Research for Social Good"

Philip Chodrow, Sharon Xu, Austin Herrling

Add to Calendar Jan/29 Mon 09:00AM-04:15PM 34-101

Enrollment: Unlimited: No advance sign-up
Prereq: None

Date: Monday, January 29th, 2018

Time: 9:30am-4:30pm

Place: 34-101

Schedule:

9:00am-9:30am

COFFEE AND REFRESHMENTS

 

 9:30am-10:15am

Michael Johnson, UMass Boston

“Community-engaged operations research: Localized interventions, appropriate methods, social impact”

 

10:30am-11:15am

Arthur Delarue and Sebastien Martin, MIT, Operations Research Center

“8 months on a school bus”

 

11:30am-12:15pm

Andrew Therriault, Chief Data Officer, City of Boston

“Saving the world with data - The case for civic data science”

 

12:30pm-1:30pm

LUNCH BREAK (not provided)

 

1:45pm-2:30pm

Edoardo Airoldi, Harvard, Statistics

 "Near-optimal design of social network experiments"

 

2:45pm-3:30pm

Hamsa Bastani, IBM Research

“Mechanism design for social good: Analysis of medicare pay-for-performance contracts”

 

3:45pm-4:30pm

Marta C. Gonzalez, UC Berkeley, Associate Professor of City and Regional Planning

“Modeling and planning urban systems with novel data sources”

 

More details available on the ORC IAP Seminar website: http://orc.mit.edu/events/orc-iap-seminar-2018

 

If you have any questions you may contact the ORC IAP Seminar student coordinators by email: orc_iapcoordinators@mit.edu

Sponsor(s): Operations Research Center
Contact: Philip Chodrow, pchodrow@mit.edu


Seven Sketches in Compositionality: Real-World Applications of Category Theory

David Spivak, Research Scientist, Brendan Fong, Postdoctoral Associate

Enrollment: Unlimited: No advance sign-up
Attendance: Participants welcome at individual sessions

Category theory is a relatively new branch of mathematics that has transformed much of pure math research. The technical advance is that category theory provides a framework in which to organize formal systems, and by which to translate between them, allowing one to transfer knowledge from one field to another. But this same organizational framework also has many compelling examples outside of pure math. In this course, we will give seven sketches on real-world applications of category theory.

No knowledge of category theory is assumed; we will build up from the basics to the advanced theory over the series of lectures. We will also provide course notes that we hope will become a book.

More details at: http://math.mit.edu/~dspivak/teaching/sp18/

Contact: Brendan Fong, 2-180, BFO@MIT.EDU


Add to Calendar Jan/09 Tue 01:30PM-03:00PM 2-255
Add to Calendar Jan/11 Thu 01:30PM-03:00PM 2-255
Add to Calendar Jan/16 Tue 01:30PM-03:00PM 2-255
Add to Calendar Jan/18 Thu 01:30PM-03:00PM 2-255
Add to Calendar Jan/23 Tue 01:30PM-03:00PM 2-255
Add to Calendar Jan/25 Thu 01:30PM-03:00PM 2-255
Add to Calendar Jan/30 Tue 01:30PM-03:00PM 2-255

David Spivak - Research Scientist, Brendan Fong - Postdoctoral Associate