Introduction In this blog post, we will explore the fundamental concepts of robotic grasping. First, we will summarize the theory and mathematical models underlying robotic grasping, focusing on t...

Introduction This will be a short post that I was initially hesitant to write, but the insights and results from this experiment are valuable enough to document. In essence, I spent about a day t...

Introduction This blog post will focus on t-SNE (t-distributed Stochastic Neighbor Embedding) as a tool for data visualization in reinforcement learning problems. We will begin by providing a quic...

Introduction Typically, an autoencoder is a type of neural network used to reduce the dimensionality of data by projecting it onto a lower-dimensional manifold. This is achieved through a network ...

Introduction When working with reinforcement learning algorithms, stability and consistency are crucial for achieving optimal performance. One powerful method, Soft Actor-Critic (SAC), combines th...

DexHand vs. LEAP Hand After evaluating various robotic hands, we’ve narrowed our choice to two contenders: DexHand and LEAP Hand Hand. Here, we provide a detailed description and comparison of bot...

Properties of Robotic Hand Projects This post aims to document and assess open-source anthropomorphic hands, with the goal of identifying the most suitable candidate for construction in our labora...

This is a summary of Julius von Kügelgen’s presentation titled “Causality for Machine Learning.” In his talk, he explored the mathematical concepts behind causality and how they differ from traditi...

Introduction In this blog post, I will delve into the paper titled “Toward Causal Representation Learning” (2021) (link). The primary objective of this paper is to address open challenges in machi...

Introduction Expanding on our previous discussions about Proximal Policy Optimization (PPO) in reinforcement learning, let’s now focus on a crucial aspect: hyperparameter optimization. In reinforc...