Paper Read 12 - Imitation Learning
Imitation Learning: An Introduction
Paper Read 11 - Transfer Learning
I stumbled upon some ideas from the paper ‘PaLM-E: An Embodied Multimodal Language Model’ that I read earlier that came from transfer learning. So I decided to do some learning in this area.
Paper Read 10 - Basic Concepts of ML
In the process of learning about transfer learning, I found myself a little fuzzy about some of the concepts of machine learning. So I decided to write a post to record the basic concepts of machine learning. I decided to record in my native Chinese language in order to make it easier to understand.
Paper Read 9 - slides: Controlling Robots via Large Language Models
A short post, as I have only started learning llm + robotics in the last few days, I am not familiar with the feasibility, difficulties and pain points of this field. I read a blog with a comparative explanation and took notes.
Paper Read 8 - PaLM-E: An Embodied Multimodal Language Model
Recently, when I saw the unmanned system of Westlake University’s explanation of Embodied Intelligence: Large Language Models Enabling Manipulators Planning and Control, it was mentioned that the intelligent capability brought by LLM makes robots smarter, more flexible, and capable of adapting to complex operation scenarios. So I read the paper of Google’s PaLM-E mentioned in the article, and the record is as follows.
Paper Read 7 - Attention Is All You Need
The emergence of LLM has advanced the field of robotics. These models can learn rich linguistic knowledge and semantic representations by being pre-trained on large-scale textual data. These models can then be fine-tuned to adapt to specific tasks or domains. Therefore, in this post, I am going to document my learning process of embodied intelligence. Since I only know about the robotics field before, and the main body of the paper is in the field of NLP, I decided to record in my native Chinese language in order to make it easier for me to understand.
Paper Read 6 - A Novel QP-Based Kinematic Redundancy Resolution Method With Joint Constraints Satisfaction
I want to document the learning process of reading a paper which uses velocity-level IK method achieving joint acceleration constraints together with the velocity- and position-level constraints.
Older
Newer