Shu IshidainTowards Data ScienceMaking LLMs Write Better and Better Code for Self-Driving Using LangPropAnalogy from classical machine learning: LLM (Large Language Model) = optimizer; code = parameters; LangProp = PyTorch LightningJun 251Jun 251
Shu IshidainTowards Data ScienceCALVIN — a neural network that can learn to plan and navigate unknown environmentsCALVIN is a neural network that can plan, explore and navigate in novel 3D environments, provided expert demonstrations.Jun 1, 2022Jun 1, 2022
Shu IshidainGeek CultureInstalling Linux (Ubuntu 20.04) on an external portable SSD and pitfalls to be aware ofCarry your Ubuntu OS anywhere without being limited by hardwareApr 22, 20224Apr 22, 20224
Shu IshidainTowards Data ScienceHow do machines plan? —An overviewA gentle introduction to graph search algorithms and Markov Decision ProcessesApr 16, 2021Apr 16, 2021
Shu IshidainTech to Inspire3 hints AI gives us about lifeHow to perform stochastic gradient ascent towards your goalsJan 30, 2021Jan 30, 2021
Shu IshidainTech to Inspire25 proverbs in AI languageTraining of a 1000 epochs begins with a single gradient descentDec 21, 2020Dec 21, 2020
Shu IshidainThe StartupHow Robots Make Maps— An Intro to SLAM (Simultaneous Localisation and Mapping)A big picture of odometry, localisation, mapping, feature matching and loop closure for non-roboticistsAug 31, 20203Aug 31, 20203
Shu IshidainTech to Inspire9 Steps to Reprogram Yourself and Stay ProductiveWe all have different ways of keeping ourselves focussed and productive. Over the course of my studies, I have seen many hardworking…Aug 14, 2020Aug 14, 2020
Shu IshidainTech to InspireHow Counting in Binary Numbers can Cause Auditory IllusionsLet’s generate Shepard Scales from a PIC microcontroller in Assembly LanguageAug 8, 2020Aug 8, 2020
Shu IshidainTech to InspireHow Assembly Language became my first programming languageThe beauty of PIC microcontrollersAug 1, 2020Aug 1, 2020