One of the major drawbacks of performing reinforcement learning, imitation learning, deep learning research on real-world environment is the availability of appropriate simulation environments. For example in reinforcement learning a game or physics simulator is needed with good fidelity to real world scenarios.

There are plenty of games and simulators out there but most often researchers are afraid to create their own environment. And to good reason: it is very hard and also time consuming to create a usable world environment and design all its functions. …

If you are a young individual about to enter the work force, congratulations! It is always an exciting time to build your life and career. But some of you may wonder what to do. The academic environment maybe is what you are used to coming out of schools. Then there are the big companies, big names you always hear about. Then there are small startup companies, which you may know less about, but that change the game…

“What shoudl I chose?” if this is your question, this may be able to help you. First some information about me: I have…

We know we are a creature of habits. We like a routine, with its well defined days and we like to be able to predict the output of the actions we perform. It has to do with the way our brain works, which inherently tries to minimize surprise.

But what about innovation, which by definition is the search for something new and not predictable, something that leaps out of the noise, of the common routine [foot note 1]? Think of a new energy source that can power houses, cars and planes without burning anything. Think of repairing the body, making…

I learn at my own pace, in my own house

In covid-19 times we had the opportunity to experiment with different learning methods beside the traditional classroom. This is good!

Having 20–30 students (or more) in a classroom listening to a live lecture is hardly an effective way to teach. There are several reasons why:

  • the time spent on getting ready, going to school is lost
  • students do not get individual attention in the classroom
  • a class is only as fast as the slowest or average student group
  • a lot of time is wasted policing behavior of multiple students and personalities

A new education

in a new coronavirus world

These are some guidelines that I would like to see governments and center for disease control set for a safe exit quarantine strategy.

In the presence of a transmissible virus we ought to change our style of lives from what we are used to, this is both painful and inconvenient, but having family member die or be compromised for life because of our laziness and selfishness is not an outcome we want to have.

In general, reopening a country after quarantine means we have to contain new spread of the virus. Lockdown at home may…

Before building, let us learn to take care

One definition of intelligence is the ability to predict the future and the results of our actions. If an outside observed was to examine how are prepared for the late 2019 — 2020 coronavirus pandemic, one may not find it to resonate well with the definition of intelligence.

But if you look closer, we did not fail to prepare or build. Rather we failed to observe, react and scale.

We have the science and knowledge to make vaccines, to make ventilators and medical equipment. We had teams of experts that know what to do and how to react. We had…

Convert all your movie files and shrink them~

If you want to save space in our hard disk, convert your movie files with the latest compression algorithms:

What are the foundation of intelligence? What can make an artificial entity capable of intelligent behavior in its own environment? Here I list the major building blocks that are necessary to learn to solve multiple tasks while acquiring knowledge and material gains. This list exists to understand where the current status of research and developement is and where we need to focus.

An intelligent agent lives in an environment that can be observed, and the agent actions have an effect on the environment (the loop)

The agent has a purpose to conduct itself in the environment

A purpose is…

We are at the end of a decade that will be forever know as the “Decade of Deep Learning and AI”. What lies ahead is exciting and will continue to change our lives.

The path of a new decade

Neural Networks, back-propagation, the promise of intelligent system, Artificial Intelligence (AI) — all have been in the making for many decades. But only with the computing power of the years after 2010, and with the large amount of datas produced by our cell-phones since the years after 2001 or so, neural networks actually became really useful in categorizing and learning from data.

Deep Learning and neural network…

This posts explain the typical sequence-to-sequence (s2s) neural networks architecture and how to use them.


Imagine we have in input sequence X= [x1, x2, x3, …] and an output sequence Y= [y1, y2, y3, y4, …]. Inputs and outputs respectively use sequences with input symbols and output symbols. These symbols can be different and often are.

Examples of s2s these are:

  • language translation: inputs sequence in one language, output in another
  • speech traslation: input are audio samples, output is text transcription
  • video description: input video frames, output is caption
  • and many more…

s2s neural network are used to translate a…

Eugenio Culurciello

I dream and build new technology

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store