top of page

ComputeFest 2021

KEYNOTE: "HOW DATA ANALYTICS AND TECHNOLOGY DESIGN WILL DICTATE OUR CIVIC FUTURE"
Presented by Prof. Latanya Sweeney, Daniel Paul Professor of the Practice of Government and Technology at the Harvard Kennedy School and in the Harvard Faculty of Arts and Sciences

​

SKILL-BUILDING WORKSHOPS:

 

The 2021 ComputeFest lectures and labs are available here: 

https://github.com/Harvard-IACS/2021-ComputeFest

​

ComputeFest 2020

KEYNOTE #1- "DATA SCIENCE: WHAT IS IT NOT?"

Presented by Xiao-Li Meng, Editor-in-Chief Harvard Data Science Review & Whipple V. N. Jones Professor of Statistics

​

KEYNOTE #2- "JUPYTER MEETS THE EARTH: AN OPEN, COLLABORATIVE APPROACH FOR EARTH DATA SCIENCE"

Presented by Fernando PEREZ, Associate Professor of Statistics, UC Berkeley & Co-founder Project Jupyter

​

SKILL-BUILDING WORKSHOPS-

  • FROM NOTEBOOK TO THE CLOUD

  • AWS SAGEMAKER

  • HOW TO USE EXISTING MODELS FOR TRANSFER LEARNING

  • TRANSFER LEARNING ACROSS TASKS: IMAGENET HELPS SEMANTIC SEGMENTATION

  • NETWORK DISTILLATION

  • AFTERNOON WORKSHOP 2 HOSTED BY WEIGHTS AND BIASES (WANDB) 

  • MODELING SEQUENTIAL DATA WITH ELMO, BERT, TRANSFORMERS, AND OTHER CHILDHOOD HEROES
  • MACHINE LEARNING FOR TIME SERIES FORECASTING WITH PYTHON
  • AFTERNOON TECH TALK SPONSORED BY AMAZON ALEXA: "THE SCIENCE BEHIND ALEXA'S ENTITY RESOLUTION"
  • UNDERSTANDING AND DIAGNOSING DEEP MODELS WITH VISUALIZATION
  • VISUALIZING NEURAL NETWORKS AND OTHER ML MODELS
  • DEEP LEARNING WORKFLOWS IN MATLAB: A HANDS-ON WORKSHOP
     

ComputeFest 2019

SYMPOSIUM - DATA SCIENCE AT THE FRONTIER OF DISCOVERY: MACHINE LEARNING IN THE PHYSICAL WORLD

 

From hydrology to earthquake prediction to cosmology, machine learning is transforming our understanding of the physical world. This year’s symposium will bring together global leaders in machine learning and computational science to discuss new approaches and advances in scientific understanding enabled by significant developments in computational power, design, and analysis.

​

​​

SKILL-BUILDING WORKSHOPS

  • INTRODUCTION TO DEEP NEURAL NETWORKS FOR IMAGE RECOGNITION

  • CONVOLUTIONAL AUTOENCODERS FOR IMAGE MANIPULATION

  • NVIDIA: FUNDAMENTALS OF DEEP LEARNING FOR COMPUTER VISION

  • INTERPRETABILITY AND FAIRNESS IN DATA SCIENCE

  • MODEL AGNOSTIC METHODS FOR INTERPRETABILITY AND FAIRNESS

  • USING GOOGLE'S WHAT-IF TOOL

  • IBM RESEARCH AI: AI FAIRNESS 360

  • NATURAL LANGUAGE PROCESSING

  • ACTIONABLE UNDERSTANDING OF CUSTOMER REVIEWS VIA CLASSIFICATION

  • NATURAL LANGUAGE PROCESSING: MICROSOFT AZURE + RESEARCH

ComputeFest 2018

SYMPOSIUM - THE DIGITAL DOCTOR: HEALTH CARE IN AN AGE OF AI AND BIG DATA


Medicine and health care, like other aspects of life in the 21st century, are being reshaped by computational science, big data, and information technology. Given gigabytes of personal genomic data, therapies can be tailored to the individual patient and the individual disease. Standardized electronic health records create an information resource that improves decision making in the doctor’s office and also sheds light on health status across entire populations. Machine learning and artificial intelligence may soon augment the physician’s diagnostic skills. As these innovations promise to improve health and prolong lives, however, they also raise sticky economic and ethical questions. Many new medical interventions come at a fabulous price; can we assure equitable access to them? How do we preserve privacy and personal dignity when the “healing arts” become a high-tech enterprise? This symposium will explore these questions and more.

​

​​

​

​

SKILL-BUILDING WORKSHOPS

  • Bayesian Inference with Stan

  • Build a Bot with Domino Data Lab

  • Deep Learning Image Classification with Keras

  • Deep Learning for Healthcare Image Analysis

  • Deep Learning with MATLAB

  • Digital Content Creation Using GANS and Autoencoders

  • Machine Learning with H20

  • Machine Learning with Microsoft AzurE

  • Machine Learning with Python sklearn

  • Pandas: Relational Database Concepts

  • Tableau

ComputeFest 2017

SYMPOSIUM - DATA DOLLARS AND ALGORITHMS: THE COMPUTATIONAL ECONOMY

​

You buy lunch from a food truck and pay by waving your cell phone; before you’ve finished your sandwich, the transaction is posted to your bank account. This is an example of how computer technology lubricates the economy. At a deeper level, computation is also essential to many aspects of financial engineering—portfolio selection, risk management, high-speed trading, the design of new market mechanisms such as online auctions, and even algorithms for the donation and exchange of human organs. With Bitcoin, money itself has become a computational object. And yet there remain pitfalls in economic life that algorithmic methods have so far failed to overcome. We still struggle to forecast and control macroeconomic cycles of boom and bust and to deal with inequities of wealth distribution. Merely measuring the state of the economy (productivity, employment, inflation) is slow and imprecise. Can abundant data and computational power play a role in improving this situation?

​

​

​

SKILL-BUILDING WORKSHOPS

  • DATA SCIENCE IN PYTHON

  • INTRO TO MATLAB

  • MATLAB AND THE INTERNET OF THINGS

  • DEEP LEARNING PART 1

  • DEEP LEARNING PART 2

  • INTRO TO TABLEAU

  • INTRO TO R

  • NOSQL VS. NEWSQL

  • R PROGRAMMING FOR DATA ANALYSIS

  • NVIDIA PASCAL 100 AND CUDA 8 DEEP DIVE

  • INTRO TO IMAGE RECOGNITION WITH MICROSOFT COGNITIVE TOOLKIT

  • MACHINE LEARNING IN THE WOLFRAM LANGUAGE

  • DEEP LEARNING FOR IMAGE SEGMENTATION WITH TENSOR FLOW

  • IMAGE PROCESSING, MACHINE LEARNING, COMPUTER VISION AND

DEEP LEARNING IN MATLAB

_IACS logo width130height130_orig.png
bottom of page