Generative meditations on one life lived and the thousands not — an ink tree and a constellation of selves

Interactive knowledge graph mapping 800+ concepts across mathematics, machine learning, and computer science

After years on Gatsby v2, I moved to Astro. Here's what worked, what didn't, and what I learned.

Are you making effective use of your customer feedback?

Generate Trajectories from the Transformers based Language Model

We visually inspect the images generated by GAN. But, with a twist.

Build CIFAR10 classifier(s) using different deep learning frameworks and hardware accelerators

Evaluation of Crowd Density Models

Counting item of interest in an image or a video is an interesting problem. Too few and you can use YOLO or SSD based object detection models, too many and you can use...

Exploration of the Global Unicorn Club

We use soft attention based technique to improve vanilla image caption model that we saw in the previous post.

This is not a galaxy cluster, then what is it?

Effect of beam width, max steps, max hypothesis on Natural Language Generation

Intuitive guide to various decoding algorithms used in natural language generation (NLG)

Generating image caption using Convolutional and Recurrent Neural Networks

Box Office Performance of Marvel Movies

Annual C02 Emissions from Fossil Fuels

The mixture density network can represent general conditional probability densities using a set of learned mixture models.

Visualise neural text generation using D3

Visualising 28 years of UST yield curve rates

In this post, we train two agents to play tennis against each other

We use deep deterministic policy gradients to solve unity ml-agent with continuous action space

We use simple policy based methods to solve CartPole environment and compare their performance

In this post, I compare the time taken to train a Convolutional neural network on native 1080ti and the accelerators available on Google Colaboratory

It can provide many useful applications but at what cost ? Privacy, mass surveillance..

Implementation of DQN to solve lunar lander environment

Temporal difference control method part 3 - Expected SARSA implementation

Temporal difference control method part 2 - SARSAMAX/Q-Learning implementation

Temporal difference control method part 1 - SARSA implementation

Output from fast style transfer

Monte Carlo control methods for finding optimal policies in reinforcement learning

Can neural networks learn to write stories ? Can they be taught to crack jokes or write articles ?

Deep dream images and video

What you looking at and where you looking at?

What do filters in a convolutional neural network learn? How do we visualize a convolutional layer? Can the Neural networks dream and if they do, do they dream of electric sheep?

Collection of images generated by DCGAN

The movie was well made but story was written poorly

The paper presents a technique to perform one shot learning with one or few examples per dataset.

Can computers learn to draw like Picasso, Vangogh, Monet or any other artist? Gatys et al. introduce an algorithm that uses a Deep neural network to generate artistic images of high perceptual quality.

My name is Barry Allen and I am the fastest man alive...

Imagine being able to learn the probability distribution of your dataset and sample from the learned distribution.

Hello world