RobertT(5d) (AGA5.67) was kind to provide enlightening analysis of the AlphaGo games on my backup relay boards:
Lee Sedol vs AlphaGo - Game 1
AlphaGo vs Lee Sedol - Game 2
Lee Sedol vs AlphaGo - Game 3
AlphaGo vs Lee Sedol - Game 4
Lee Sedol vs AlphaGo - Game 5
Separately, Demis Hassabis gave a 1 hour presentation on DeepMind and AlphaGo 2 weeks before the matches. Unexpectedly detailed description on AlphaGo background and structure:
Artificial Intelligence & the Future by Dr.Demis Hassabis
Press Conference after Game 1
Press Conference after Game 2
Press Conference after Game 3
Press Conference after Game 4
Press Conference after Game 5
Go professionals are not AI experts, and AI experts are not pro-level Go experts. I have read many reviews on this game series, and found this one by Zhe Li, 6p to be striking a good balance. Articles published in China, translated by Prudential Center Go Club in Boston:
Game 1 Lee Sedol’s Strategy and AlphaGo Weakness
Game 2 Nobody Could Have Done a Better Job than Lee Sedol
Game 3 and 4
Game 5 translation work under way
Who is Zhe Li, 6p
Some pictures of David Silver’s presentation on indicated Winrate (winning probability percentage) by AlphaGo version 18 during the games: https://www.reddit.com/r/baduk/comments/4cgitz/alphagos_confidence_chart_for_the_5_games_david/