CodeIsGo.com

By optimizing reinforcement-learning algorithms, DeepMind uncovered new details about how dopamine helps the brain learn. By Karen Hao.

In 1951, Marvin Minsky, then a student at Harvard, borrowed observations from animal behavior to try to design an intelligent machine. At the time, neuroscientists had yet to figure out the mechanisms within the brain that allow animals to learn in this way.

In a paper published in Nature, DeepMind, Alphabet’s AI subsidiary, has once again used lessons from reinforcement learning to propose a new theory about the reward mechanisms within our brains.

At a high level, reinforcement learning follows the insight derived from Pavlov’s dogs: it’s possible to teach an agent to master complex, novel tasks through only positive and negative feedback.

It turns out the brain’s reward system works in much the same way—a discovery made in the 1990s, inspired by reinforcement-learning algorithms. Read more to learn the study outcomes!

Yohan Lasorsa published this guide about how to approach your open source project development. If you plan to maintain an open-source project in the long run, what you must do is automate.

The worst way of losing time is when your automated builds are not working as intended: flakiness, unreliable builds, and cross-platforms issues can (will) drive you crazy. Pragmatism should be your main driver here, seek to get the job done for the least time invested.

Azure Pipelines gives you quite a lot for free compared to other solutions. It’s fast (offers 10 parallels workers per organization for open source tier).

The article then describes:

• How to setup Azure Pipelines for an open-source project hosted on GitHub
• Requirements, how to prepare your repository
• Create the Azure DevOps project
• Setup the pipeline, yaml config
• Adding a build matrix

CI is one of the first things you should put in place in any open source project to simplify your life as a maintainer. Good read!

Mark Noonan posted this article about what pseudo-elements are and how to use them, as well as look at some common use cases where these pseudo-elements can make life easier for us.

::before and ::after are great for decorations like custom bullet points, arrows, or other features that should go with an HTML element, but are not part of the main content of the element. Often you can achieve the same result by adding extra HTML elements to your markup, and pseudo-elements are a way to avoid clutter.

The article content is split into:

• First, why are they called pseudo-elements?
• Always set the content property
• What are these pseudo-elements used for and why?

Unlike most other elements, ::before and ::after don’t appear in the HTML code for a page.

Just remember, there are many possible uses of pseudo-elements available, and you can learn more in Kevin Powell’s awesome video series about them on YouTube. Nice one!

An article by by Leon Stigter about serverless development model and how that model works on AWS Fargate, which allows you to run containers without having to manage servers or clusters.

Serverless is a development model where developers focus on a single unit of work and can deploy to a platform that automatically scales, without developer intervention.

The article has this sections:

• Containers, like really?
• Where do your containers come from?
• What does that cost me?
• Comparing the results

Deployment scripts for AWS Faragate available in GitHub repository.

Note: there is a tradeoff between control and abstraction and as a developer you have to decide if and how you want to make that tradeoff. Nice read!

Jason Brownlee wrote this tutorial to get you started with Deep Learning in TensorFlow. Predictive modeling with deep learning is a skill that modern developers need to know.

TensorFlow is the premier open-source deep learning framework developed and maintained by Google. Although using TensorFlow directly can be challenging, the modern tf.keras API beings the simplicity and ease of use of Keras to the TensorFlow project.

The article deals with:

• The difference between Keras and tf.keras and how to install and confirm TensorFlow is working
• The 5-step life-cycle of tf.keras models and how to use the sequential and functional APIs
• How to develop MLP, CNN, and RNN models with tf.keras for regression, classification, and time series forecasting
• How to use the advanced features of the tf.keras API to inspect and diagnose your model
• How to improve the performance of your tf.keras model by reducing overfitting and accelerating training

In this tutorial, you discovered a step-by-step guide to developing deep learning models in TensorFlow using the tf.keras API. Code and links to further resources, including some data sets also included. Well done!

Visualizing the structure of your neural network is quite useful for publications, such as papers and blogs. Written by @MachineCurve (Christian).

Today, various tools exist for generating these visualizations – allowing engineers and researchers to generate them either by hand, or even (partially) automated.

In this blog post, we’ll take a look at Net2Vis. Firstly, we’ll inspect the challenges of current tools in more detail, followed by the introduction of Net2Vis. We then suggest a different way of installing it, by using our Docker-based installation process, saving you quite some time on installing dependencies. Subsequently, we’ll talk you through our experience with Net2Vis – and show you what it’s capable of.

The article then walks over these topics:

• What is Net2Vis?
• Using Net2Vis online
• How does Net2Vis work?

Often, print media requires horizontal visualizations, maintaining the natural flow of reading, while still conveying all important information. Very useful for anybody interested into machine learning. Nice one!

Kick the tires with faasd today, for a lightweight serverless experience that doesn’t require Kubernetes. Instead it uses containerd and the existing OpenFaaS ecosystem. By Alex Ellis.

You can run faasd anywhere, even on a Raspberry Pi, but why would you want to do that? faasd offers many of the benefits of containers and OpenFaaS, but without the complexity and operational costs of Kubernetes. containerd is a low-level tool for automating containers, and a CNCF project.

Compute is compute, and OpenFaaS with containerd makes it easy to both consume and provide.

The article briefly covers:

• Before we start and installation
• Post-install
• How to build a new container image
• Deploy your new container

faasd brings a lightweight experience to your Raspberry Pi, cloud infrastructure and to bare metal, all without the need for Kubernetes. Nice read!

An in-depth guide to some important console API functions other than console.log. By Deepak Gupta.

The Console API’s most commonly used method is undoubtedly console.log. However, it also comes with several other useful methods that can improve our debugging efficiency.

The article describes some other interesting methods:

• console.table(data[, properties])
• console.group(label) & console.groupEnd(label)
• console.dir(object) & console.dirxml(object)
• console.profile([, label]) & console.profileEnd([, label])
• console.time([, label]) & console.timeEnd([, label])

… and more. All described here methods are available in NodeJS as of version 10.0.0 and in almost every major browser. For the rest the link to original article. Nicely put together!

Carlos Caballero published this story about fundamental tips and advice to generate clean code focusing on the element that allows us to reuse our code: the functions.

Some tips to generate good code applied specifically to the variables:

• Use default arguments instead of short circuiting or conditionals
• Function arguments (2 or fewer ideally)
• Avoid side effects – Global Variables
• Avoid side effects – Objects Mutables
• Functions should do one thing
• Favor functional programming over imperative programming
• Use method chaining

The design of functions applying clean code is essential because the functions are the basic element to decouple the code. Good read!

This article is part of Demystifying AI, a series of posts that (try to) disambiguate the jargon and myths surrounding AI. By Ben Dickson.

Before becoming too excited about advances in computer vision, it’s important to understand the limits of current AI technologies. While improvements are significant, we are still very far from having computer vision algorithms that can make sense of photos and videos in the same way as humans do.

The article focuses on:

• Commercial applications of computer vision
  • Image search
  • Image editing and enhancement
  • Facial recognition applications
  • Data efficient home security
  • Interacting with the real world
• Advanced applications of computer vision
  • Medical image processing
  • Playing games
  • Cashier-less stores
• Creepy applications of computer vision
  • Surveillance
  • Autonomous weapons

… and much more. Worth to notice that renowned computer scientist and AI researcher Stuart Russell has founded an organization dedicated to stopping the development of autonomous weapons. Good read!