On-Policy vs. Off-Policy Monte Carlo, With Visualizations

In Reinforcement Learning, we either use Monte Carlo (MC) estimates or Temporal Difference (TD) learning to establish the ‘target’ return from sample episodes. Both approaches allow us to learn from…

medium.com_towards-ai 2 weeks ago

This AI newsletter is all you need #63

This week in AI, we highlight further evidence of the adoption of Large Language Models (LLMs) driving significant revenue growth in the Western market, as well as the introduction of new AI models…

medium.com_towards-ai 2 weeks ago

Two Ways to Download and Access Llama 2 Locally

Meta’s latest release, Llama 2, is gaining popularity and is incredibly interesting for various use cases. It offers pre-trained and fine-tuned Llama 2 language models in different sizes, from 7B to…

Power of Context in Data-Driven Storytelling

Data-driven storytelling is a way of communicating numerical information through narration and visualization. Its goal is to engage the audience and help them better understand the main conclusions…

Environmental Data Science: An Introduction

As interest in sustainability is growing, environmental data science comes into focus. We show examples, outline challenges and perspectives.

Reinforcement Learning: SARSA and Q-Learning — Part 3

In the previous article — Part 2 — we discovered a few solution algorithms to solve the Markov Decision Process (MDP), namely the Dynamic Programming method and the Monte Carlo method. The Dynamic…

medium.com_towards-ai 2 weeks ago

Wild Wild RAG… (Part 1)

RAG (Retrieval-Augmented Generation) is an AI framework that enhances the quality of Language Model-generated responses by incorporating external knowledge sources. It bridges the gap between…

medium.com_towards-ai 2 weeks ago

Microsoft Research Introduces Not One, Not Two, But Four New AI Compilers

Compilers are seeing a renaissance in the era of generative AI. In the context of AI, a compiler is responsible for translating a neural network architecture into executable code in a specific…

medium.com_towards-ai 2 weeks ago

Machine Learning: Understand Centering and Scaling purposes

This article introduces the centering and scaling concepts. With a real-world use case, I explain the advantages of the center and scale the data. Technically, we compare the MinMaxScaler…

medium.com_towards-ai 2 weeks ago

Training an Agent to Master a Simple Game Through Self-Play

Isn’t it amazing that everything you need to excel in a perfect information game is there for everyone to see in the rules of the game? Unfortunately, for mere mortals like me, reading the rules of a…