Data Science et al.

Regular Expressions

Reinforcement Learning

Statistics

Tensorflow vs Numpy

Zero Shot Learning

Will AutoML steal the jobs of Deep Learning engineers and practitioners?

Support the show (http://paypal.me/SachinPanicker )

Robust Statistics

Support the show (http://paypal.me/SachinPanicker )

Deep Reinforcement Learning

Support the show (http://paypal.me/SachinPanicker )

AutoGluon - 1

Introducing AutoGluon-Tabular, an open source AutoML framework Support the show (http://paypal.me/SachinPanicker )

Guest Episode - The Music and Vibes Podcast

This is a Guest special episode, featuring Keana W. Mitchell, a relationship consultant living in Madison, AL. She has a Master's Degree in Counseling Psychology and has worked for eight years for the state of AL as a Rehabilitation Counselor. She is also a singer, songwriter, and podcaster who is very passionate about helping couples improve their marriage through counseling, education, and music. She is currently in the process of starting her own relationship consulting firm called "The E...

Statistics - Descriptive vis-à-vis Inferential

Types of Statistics explained Support the show (http://paypal.me/SachinPanicker )

Python or R?

Which programming language serves data scientists well? Support the show (http://paypal.me/SachinPanicker )

Feature Subset selection

How does one select a feature subset? Support the show (http://paypal.me/SachinPanicker )

Genetic Algorithm

GA has been widely applied in path optimization problem which consists in finding a valid and feasible path between two positions while avoiding obstacles and optimizing some criteria such as distance (length of the path), safety (the path must be as far as possible from the obstacles) etc.  Support the show (http://paypal.me/SachinPanicker )

Neptune - 1

The 1st in a series of 7 episodes on Neptune. Support the show (http://paypal.me/SachinPanicker )

RegEx

Usage of RegEx in ML Support the show (http://paypal.me/SachinPanicker )

TensorFlow or NumPy

A sneak peek into the advantages of using TensorFlow. Support the show (http://paypal.me/SachinPanicker )

Encoder

In this episode, we take a look into some of the encoders used to encode categorical features in Machine Learning. Support the show (http://paypal.me/SachinPanicker )

How to build an AI Product?

The seven essential steps involved in building an AI product. Support the show (http://paypal.me/SachinPanicker )

Feature Subset

What is feature subset selection? Support the show (http://paypal.me/SachinPanicker )

Episode 36 - Helix - 1

HELIX is a machine learning system that optimizes the execution across iterations—intelligently caching and reusing, or recomputing intermediates as appropriate. Support the show (http://paypal.me/SachinPanicker )

Episode 35 - Apache Arrow

Introducing Apache Arrow - an in-memory data structure specification. Support the show (http://paypal.me/SachinPanicker )

Episode 34 - Michelangelo

Introducing Michelangelo - Uber's machine learning platform Support the show (http://paypal.me/SachinPanicker )

Episode 33 - Hidden Markov Model

A multi layer Hidden Markov Model can be very effective in detecting intrusions in 5G Networks, as such networks are more vulnerable as they are IP based. Support the show (http://paypal.me/SachinPanicker )

Episode 32 - Markov Decision Process

Markov Decision Process and Markov Reward Process explained. It has immense use in the field of Reinforcement Learning. Support the show (http://paypal.me/SachinPanicker )

Episode 31 - Generative Teaching Network

GTNs allow researchers to quickly evaluate the learning potential of a newly proposed candidate architecture that catalyzes the search for new and powerful neural network architectures. Support the show (http://paypal.me/SachinPanicker )

Episode 30 - Polynote

Netflix has open sourced its Polynote, a new polyglot notebook with first class Scala support. Support the show (http://paypal.me/SachinPanicker )

Episode 29 - Manifold

Uber's model agnostic visual debugging tool for Machine learning.

Episode 28 - Horovod

Horovod, developed by Uber, and hosted by the Linux foundation, is available under the Apache 2.0 license.

Episode 27 - Flyte - 1

This is the first in a series of 5 episodes dedicated to Flyte, which was recently open sourced by the ridesharing company, Lyft.

Episode 27 - Flyte

This episode is dedicated to Flyte, which was recently open sourced by the ridesharing company, Lyft.

Episode 26 - Ludwig - 4

This is the fourth in a series of 4 episodes dedicated to Ludwig, Uber's open sourced deep learning toolbox. This covers the current limitations of Ludwig and what can be targeted in future for enhancements.

Episode 25 - Ludwig - 3

This is the third in a series of 4 episodes dedicated to Ludwig, Uber's open sourced deep learning toolbox.

Episode 24 - Ludwig - 2

The Encoders Combiner Decoders Architecture !

Episode 23 - Ludwig - 1

This is first in a series of 4 episodes dedicated to Ludwig, Uber's open sourced deep learning toolbox.

Episode 22 - PyText - 3

This is the third and final episode in a series of 3 episodes dedicated to PyText, in which we explore the stuff that PyText comprises of internally.

Episode 21 - PyText - 2

In this episode, we unravel PyText further, which is a deep learning based NLP modeling framework built on PyTorch. Also we compare other NLP frameworks - CoreNLP, AllenNLP, FLAIR, Spacy 2.0

Episode 20 - PyText - 1

In this episode, we unravel PyText, which is a deep learning based NLP modeling framework built on PyTorch.

Episode 19 - Best practices for VR - 2

As a VR App developer, it is your obligation to Users to provide them a captivating and enchanting experience while also ensuring their comfort and safety.

Episode 18 - Best practices for VR - 1

The best practices to be considered for developing a Virtual Reality app include, but not limited to the following aspects - User experience User input Vision Audio Locomotion Avatar Rendering User orientation and positional tracking

Episode 17 - Zero Shot Learning

Facebook is doing some interesting work in the field of Generative Adversarial Zero Shot Learning. Another of their research project LASER uses this technique to understand multiple dialects and languages by learning similarities which are subtle and not discernible to humans!

Episode 16 - Evaluation of language models

Extrinsic evaluation  • Is the evaluation on a real task  • Can be slow to compute performance Whereas Intrinsic evaluation  • Is the evaluation on a specific, intermediate task  • Fast to compute performance  • Helps understand subsystem

Episode 15 - Basics of NLP

Text Normalization and sentence segmentation

Episode 14 - MLflow

MLflow provides three components to help manage the ML workflow: MLflow tracking MLflow projects MLflow models

Episode 13 - Regular Expressions

RegEx - The very basics of Natural Language Processing

Episode 12 - TensorFlow or NumPy

Advantages of TensorFlow - Computation Graphs

Episode 11 - Edge Intelligence

The revolution in AI

Episode 10 - Encoders in ML

Episode 8 - How to build an AI product?

Episode 9 - Probability distributions