Moving from prototypes to products: How to build and deploy at-scale Data Science driven products

location_city Bengaluru schedule Sep 1st 02:00 - 02:45 PM place Grand Ball Room 2 people 75 Interested

Have you ever wondered what it takes to demonstrate the efficacy of a data-driven approach to solve a particular problem and then build it into a full-fledged product that can cater at the big-data scale? Although frameworks like Six Sigma exist for software development life cycle, there are no standard best practices for data science driven product development and continuous improvement lifecycle. There is tremendous amount of literature available on internet regarding the machine learning model training and inference, yet there is a lack of publicly accessible know-how on how to structure machine learning projects.

At Envestnet | Yodlee, we have deployed several advanced state-of-the-art Machine Learning solutions which process millions of data points on a daily basis with very stringent service level commitments.

In this session, we will elaborate our framework for developing and deploying data-science driven products in the context of our latest product: Transaction Data Enrichment (TDE). TDE caters to millions of financial transaction requests daily which originate from thousands of different sources. Learnings from this ML-driven product deployment will help ML practitioners and data science leaders on how to structure their ML projects, how to build an optimal prototype, and how to quickly scale from the prototype stage to the production deployment stage.

Specifically, we will talk about best practices and challenges for each stage of the data driven product development starting from (a) data gathering, cleaning and normalization, (b) incorporating domain expertize, (c) choice of model, (d) model training and evaluation, (e) design decisions for at-scale deployment, (f) planning for redundancy, (g) continuous monitoring for model deterioration, and (h) logging customer feedback

With the latest advancements in machine learning methods like deep learning and the availability of vast amount of data, there has never been a better time to build data science driven products. Come join us as we share our learning in this journey.


Outline/Structure of the Case Study

  • Where to apply machine learning and how to formulate a problem statement
  • Best practices for getting your data tagged for supervised models
  • How to create a representative train and test set for model training and evaluation
  • Which ML framework / DL model architecture best suits your problem
  • Best engineering practice for quickly optimizing hyper-parameters of deep learning model architecture

Learning Outcome

  • You will learn how to prioritize your efforts on different directions in ML project
  • You will get know importance of error analysis and the virtuous loop continuous training and model improvement
  • You will learn best practices about evaluating your machine learning model

Target Audience

People interested in machine learning and data driven product development and deployment

Prerequisites for Attendees

Basic knowledge of machine learning and familiarity with deep learning.


schedule Submitted 3 years ago

Public Feedback

    • Samiran Roy

      Samiran Roy / Dr. Om Deshmukh - Reinforcement Learning: Demystifying the hype to successful enterprise applications

      45 Mins
      Case Study

      In 2014, Google accquired DeepMind, a small, london-based AI startup for $500 million. DeepMind was conducting research on AI that would learn to play computer games in a fashion similar to humans. In 2015, Deepmind published a paper in Nature, describing a learning algorithm called Deep-Q-Learning which was able to achieve superhuman performance on a diverse range of Atari 2600 games[1]. They achieved this without any domain specific engineering - The algorithm took only the raw game images as input, and was guided by the game score. Believed by many to be the first steps in Artificial General Intelligence, DeepMind achieved this by pioneering the fusion of two fields of research - Reinforcement Learning(RL) and Deep Learning.

      RL is a learning paradigm inspired by operant conditioning which closely mimics the human learning process. It shifts focus from ML based pattern recognition solutions to learning through trial and error via interaction with an environment, guided by a reward signal or reinforcement. Imagine an agent teaching itself how to steer by navigating the streets of Grand Theft Auto - and transferring this knowledge to a driverless car[2]. Think of team of autonomous robots collaborating to outwit their opponents in a game of Robot Soccer[3]. Any practical real-world application suffers from the curse of dimensionality (A camera mounted on a robot feeding it a 64*64 grayscale image will have 256^(4096) input possibilities). A Deep Neural Network automatically learns compact and efficient feature representations from noisy, high-dimensional sensory inputs in its hidden layers, giving RL algorithms the edge to scale up and give incredible results in dynamic and complex domains.

      The most notable example of this is AlphaGo Zero[4] - the latest version of AlphaGo, the first computer program to defeat a world champion at the game of Go (Also called Chinese Checkers). AlphaGo Zero uses RL to learn by playing games against itself, starting from completely random play, and quickly surpasses human expert performance. Not only is the game extremely complex (A 19*19 Go board can represent 10^170 states of play), accomplished Go players often struggle to evaluate whether a certain move is good or bad. Most AI researchers were astonished by this feat, as it was speculated that it would take atleast a decade for a computer to play Go at an expert human level.

      RL, which was largely confined to academia for several decades is now beginning to see some successful applications and products in the industry, in fields such as robotics, automated trading systems, manufacturing, energy, dialog systems and recommendation engines. For most companies, it is an exciting prospect due to the AI hype, but very few organizations have identified use cases where RL may play a valuable role. In reality, RL is best suited for a niche class of problems where it can help automate some tasks(or augment a human expert). The focus of this presentation will be to give a practical introduction to the RL Setting, how to formulate problems into RL, and presenting successful use cases in the industry.