A3C: Asynchronous Advantage Actor-Critic A3C is a deep reinforcement learning method that consists of mainly 3 elements. Element 1: Asynchronous Instead of only having one agent trying to get to the desired destination, this paper has multiple agents exploring the…
Eligibility Trace Let’s say we want to use Q-Learning(a reinforcement learning method) to have an agent get from point A to a certain destination. Without Eligibility Trace, the agent will take one step and get feedback as a “REWARD” to…
Intuition Deep Convolutional Q-Learning is a reinforcement learning method that visually perceives an image to understand what to do next to maximize the reward to achieve a certain task. Let’s say we are playing Mario. We can use DCQLearning to…
ONNX Due to the increase of frameworks for developing machine learning models, working across frameworks and hardware ecosystems became extremely difficult. ONNX (Open Neural Network eXchange) is an open-source standard that defines a common set of operators and a file…
NNs and GNNs NN’s Expects inputs as a node GNN’s Inputs/Outputs graphs instead of a singular node. Expects inputs as an array of nodes. Considers “Adjacency Matrix” which stores information on the relationships between each node, besides the results from…
Findings for Tabular Medium-sized Datasets While deep learning has enabled tremendous progress on text and image datasets, results show that tree-based models remain state-of-the-art on tabular medium-sized data(~10K samples). Finding 1: NNs are biased to overly smooth solutions. NN’s struggle…
Otsu Binarization Image thresholding is used during the preprocessing phase to binarize images depending on the pixel intensities. Otsu Binarization is one method to find the appropriate threshold by minimizing the variance between each class. Here is an implementation example…
The 6 Phases CPMAI is a Best-practice methodology for AI Projects proposed by Cognilytica. This methodology consists of 6 phases. Phase 1: Business Understanding Assess the business situation Should you even use an AI in the first place? Is it…
4 Categories There are mainly 4 Image generation methods. 1. GAN Train a GENERATOR that generates an image from “z”, and have a DISCRIMINATOR discriminate whether the generated image is real or not. The GENERATOR tries to learn to be…
CVAT Here is one tool you can use to create annotation data for free. CVAT: – You can annotate data for many computer-vision-related tasks(semantic segmentation, 3d object detection, etc.) – You can export with multiple format options Link: CVAT