A man with a damaged emotional brain kept his full IQ, but lost the ability to choose a restaurant. This is the real case that proves feeling isn't the enemy of intelligence, it's the engine of it.
This video traces 60 years of reinforcement learning, the technology behind every AI that has mastered a game, controlled a robot, or learned to act without being explicitly programmed. It starts with 300 matchboxes in 1961 and ends with robots learning to imagine consequences before acting.
00:00 – Why feeling is the engine of intelligence
2:00 – 300 matchboxes and the birth of machine learning
4:00 – The value function: how machines learn to judge
6:00 – Samuel's checkers and the first self-improving machine
8:00 – TD-Gammon and the neural network breakthrough
10:00 – Deep Q-learning and the Atari breakthrough
12:00 – From games to robots: the physical intelligence problem
If this changed how you think about AI, hit subscribe, this channel breaks down the ideas actually driving the field, not just the headlines. Drop a comment: which system surprised you most, TD-Gammon or the Atari agents?
Instagram: https://www.instagram.com/theunseen.change/
This video is for educational purposes. All claims are based on published research cited within the video.
#reinforcementlearning #artificialintelligence #machinelearning #ai #deepmind12 #openai #aiexplained #howaiworks #neuralnetworks #qlearning #robotics #technology #aihistory #deeplearning #futureoftech
Дата на публикация: 16 октомври, 2026
Категория:
Друго
Харесайте ни във Facebook