Title:Real World Reinforcement Learning in Action
Date:9:00am-12:00pm, Oct. 10th, Thursday
Instructor:Tyler Clintworth and Microsoft Team
Outline: Microsoft recently announced the Azure Cognitive Service, Personalizer, aimed at democratizing real world reinforcement learning for content personalization. Its goal is to make reinforcement learning accessible to everyone, not just machine learning experts. Personalizer is the result of a successful partnership between Microsoft Research and Azure Cognitive Services aimed at rapid technology transfer and innovation. In this workshop you will learn the theory behind reinforcement learning, contextual bandits and how this applies to content personalization. We will walk you through setting up the service, writing your first application, and optimizing the policy using offline optimization.
  • 9:00am: Introduction to reinforcement learning and contextual bandits
  • 9:45am: Overview of Azure Cognitive Services Personalizer
  • 10:15am: break
  • 10:30am: Hands on: Setting up SDK and writing first application
  • 11:10am: Hands on: Counterfactual evaluation and offline policy optimization
  • 11:45am: Wrap-up and Q&A
  • Who should learn:Developers, Data Scientists who are working on machine learning, deep learning.
    Difficulity Level:Beginner to Intermediate
    Prerequsite:* Python and basic knowledge on machine learning
    * Bring laptop (and charger)
    Back to Home
    Title:AI at Any Scale
    Date:1:00pm-4:00pm, Oct. 10th, Thursday
    Instructor:Ion Stoica, Robert Nishihara, and Philipp Moritz, UC Berkeley
    Outline: Reinforcement learning (RL) is emerging as a promising approach to intelligently interact with continuously changing physical or virtual environments. Many see RL growing rapidly into a potentially dominant area in ML over the next decade. However, the applications of RL pose a new set of requirements, the combination of which creates a challenge for existing distributed execution frameworks: computation with millisecond latency at high throughput, adaptive construction of arbitrary task graphs, and execution of heterogeneous kernels over diverse sets of resources. Ray is a new distributed execution framework for reinforcement learning applications. We lead a deep dive into Ray, walking you through its API and system architecture and sharing application examples, including several state-of-the-art RL algorithms.
  • 9:00am: Overview of reinforcement learning and Ray
  • 10:00am: Scaling up hyperparameter search and RL
  • 11:00am: Scaling up machine learning
  • 11:45am: Wrap-up and Q&A
  • Who should learn:Developers, Data Scientists who are working on machine learning, deep learning.
    Level:Beginner to Intermediate
    Prerequsite: * Python and basic knowledge on machine learning
    * Bring laptop (and charger)
    Back to Home
    Title:Serverless Machine Learning with TensorFlow
    Date:9:00am-4:00pm, Oct. 11th, Friday
    Instructor:Lak, Tech Lead from Google.
    Lak, is a Tech Lead for Data and Machine Learning at Google Cloud where he enables customers of Google Cloud to be successful with Big Data and Machine Learning products through courses, consulting, evangelism, and open-source examples. He is the author of Data Science on GCP (O'Reilly) and Machine Learning on GCP (Coursera). He founded Google's Advanced Solutions Lab ML Immersion program. Before Google, Lak was a Director of Data Science at Climate Corporation and a Research Scientist at NOAA. Follow him on Twitter at @lak_gcp, read articles by him on Medium, and see more details at www.vlakshman.com
    Course Outline: This workshop provides a hands-on introduction to designing and building machine learning models on structured data on Google Cloud Platform. You will learn machine learning (ML) concepts and how to implement them using both BigQuery Machine Learning and TensorFlow/Keras. You will apply the lessons to a large out-of-memory dataset and develop hands-on skills in developing, evaluating, and productionizing ML models.
    Outline: (each of module is about 45 minutes and consists of 15 min lecture and 30 min codelab)
    • Module 0: Intro to machine learning and identify problems can be solved by ML
    • Module 1: Intro to BigQuery, TensorFlow and Cloud AI Platform Notebooks
    • Module 2: Use BigQuery ML to build our first ML models for taxifare prediction
    • Module 3: Learn how to read large datasets using TensorFlow.
    • Module 4: Build a DNN model using Keras.
    • Module 5: Improve the basic models through feature engineering
    • Module 6: Carry out equivalent feature engineering in Keras
    • Moudel 7: Productionize the models
    Who should learn:Developers, Data Scientists who are working on machine learning, deep learning.
    Level:Beginner to Intermediate
    Prerequsite:The following is prefered but not required.
    • Experience using Python
    • Basic proficiency with a common query language such as SQL
    • A working knowledge of data modeling and extract, transform, load activities
    • Basic familiarity with machine learning and/or statistics
    • join slack for discussion
    Back to Home