Articles

2 Things You Need to Know About Reinforcement Learning: Computational and Sample Efficiency

,

The High Cost of Deep Learning

Have you ever put on a sweater because the air conditioning was too cold? Forgotten to turn off the lights in another room before heading to bed? Do you commute to work more than 30 minutes every day just for the sake of “filling seats” at the office, even though everything you do at work could be done via laptop from home? 

In the counter-intuitive trade-offs between sample and computational efficiency in Reinforcement Learning, choosing evolution strategies can be smarter than it looks.

Source de l’article sur DZONE

/par

TensorFlow vs. PyTorch vs. Keras for NLP

,

Before beginning a feature comparison between TensorFlow, PyTorch, and Keras, let’s cover some soft, non-competitive differences between them.

Non-competitive facts:

Below, we present some differences between the 3 that should serve as an introduction to TensorFlow, PyTorch, and Keras. These differences aren’t written in the spirit of comparing one with the other but with a spirit of introducing the subject of our discussion in this article.

Source de l’article sur DZONE

/par

Light Switch With Natural Language Interface

,

In this article, I’ll show how to build a natural language interface for a typical home light switch so that you could turn the lights on and off with simple commands like Turn off all the lights, please, or Get the lights on in the kids bedroom.

We’ll concentrate on Natural Language Interface (NLI) part, and I’ll leave speech-to-text and the actual light controlling outside of the scope of this short blog. You can easily add speech-to-text with WebSpeech, if necessary, while Arduino/HomeKit can provide simple API to control the lights in your house.

Source de l’article sur DZONE

/par

Everything You Need to Know About Building an OCR Scanner from Scratch

, ,

Optical Character Recognition (OCR) tools have come a long way since their introduction in the early 1990s. The ability of OCR software to convert different types of documents such as PDFs, files, or images into editable and easily storable format has made corporate tasks effortless. Not only this, it’s ability to decipher a variety of languages and symbols gives Infrrd OCR Scanner an edge over ordinary scanners.

However, building a technology like this isn’t a cakewalk. It requires an understanding of machine learning and computer vision algorithms. The main challenge one can face is identifying each character and word. So in order to tackle this problem we’re listing some of the steps through which building an OCR scanner will become much more clearer. Here we go:


Source de l’article sur DZONE (AI)

/0 Commentaires/par

[DZone Research] Popular AI Algorithms and Their Uses

, ,

This article is part of the Key Research Findings from the DZone Guide to Artificial Intelligence: Automating Decision-Making.

Introduction

As part of the research for our 2018 Guide to Artificial Intelligence, we surveyed 403 developers, data scientists, and technologists. From their responses, we’ve created a quick article on the different algorithms available for working with various AI projects, and which one proved more popular.


Source de l’article sur DZONE (AI)

/0 Commentaires/par

Start NLP with a Single Neuron

, ,

What Is a Classification Problem?

Classification is an important and central topic in ML, which has to do with training machines how to group together data by particular criteria. Classification is the process where computers group data together based on predetermined characteristics — this is called supervised learning. There is an unsupervised version of classification, called clustering where computers find shared characteristics by which to group data when categories are not specified.

For example:


Source de l’article sur DZONE (AI)

/0 Commentaires/par

Natural Language Interface In the Enterprise

, ,

Why Enterprise Application Companies Should Take a Cue From Apple’s Siri and Google Assistant

Enterprise applications are the next frontier in the adoption of natural language interfaces. Unlike consumer tech, e-commerce, and various chatbots where NLP/U is more of a technical novelty, the world of enterprise is a killer ground for natural language interfaces.

A Need for a Unified Interface

One of the key unique properties of natural language is the fact that it provides a unified interface to any data source or sources. It’s the one interface that everyone already knows, and at the same time, it’s the same interface to any supporting system. Think about it…you can easily ask a lawyer, salesman, or marketing professional about any specific topic as long as you can formulate a question in a minimally understandable way.


Source de l’article sur DZONE (AI)

/0 Commentaires/par

The Problem With Conversational Interfaces: The Conversation

, ,

This title probably looks contrarian at a glance (so is my last post), but I truly believe we are largely misunderstanding what a natural language interface to our applications should look like. Here are my thoughts on the role of conversation in NLU/P systems.

What Is the Conversation?

Let’s define what we mean by conversation in the context of NLU/P systems. First off, conversation happens between two or more participants (computers talking to themselves at night is outside of the scope of this blog). Second, the conversation is a sequence of two or more sentences that are tightly coupled to each other by their context and time.


Source de l’article sur DZONE (AI)

/0 Commentaires/par