This past weekend our very own Kostiantyn Isaienkov, a data science and machine learning engineer, had the opportunity to speak at “Octopus AI”. Octopus AI is a data science conference which has the goal of strengthening professional skills in the world of artificial intelligence, machine learning, and data science, according to the conference website. The conference took place in the Ukraine, and welcomed professionals from all over the world.
Kostiantyn hosted a workshop on dimension reduction for image clustering using convolutional auto-encoder where he was able to showcase his vast knowledge on the subject of artificial intelligence. During this workshop, Kostiantyn presented a solution that allows engineers to compress image into a linear space which, in turn, represents the image as a vector. This use of compression allows images to cluster in a simpler fashion and can also speed up the training model process while increasing the rate of accuracy.
During the workshop, Kostiantyn presented his solution in four steps:
1. Build and train auto-encoder model
2. Dimensionality reduce real image datasets via auto-encoder model
3. Compare precision of convolutional neural network using original dataset and compressed dataset
4. Cluster image dataset
Way to represent Akvelon, Kostiantyn! Thank you to our friends at the Octopus AI conference for giving our company a chance to show our peers what Akvelon is all about.