Episode 178 – L33T Learning Siri

 

Just say no to enabling autonomous bots & drones with the ability to apply lethal force.  Following on last week’s episode 177, Michael & Michael start off with a continuation of the discussion on the Pandora’s Box of armed robots.

Moving on to a happier topic, while still staying on the machine learning concept, the pair talks about how Siri and other voice interactive systems have been improved with deep learning.  Not just recorded scripts, but actual intelligence, and more natural interaction is apparent when you listen to how far things have come since the earliest days of Siri — listen to the examples from the Apple Machine Learning Journal article, and we think you’ll agree.

Understanding that you’re talking with a bot vs a human is important to help avoid the uncanny valley where the interaction becomes weird vs expected — perhaps this is why Michael R preferred the Australian accent reminiscent of J.A.R.V.I.S, and how in Star Trek and other science fiction, interaction with the computer was prefaced with “computer!” before issuing a request.

Feedback loops getting faster & faster, which is really evident in machine learning examples applied to games.  When the machine — the program — is experimenting with the game to learn, and figure out what is rewarded, what is not rewarded, and determine a heuristic model for playing the game, which for a human is for fun and for the machine learning example is to maximize the outcome.  The example from the O’Reilly article was particularly intriguing to Michael M who wondered why the AI playing the game did not scoop up each and every present while playing the game.  Michael R made the interesting point that if the machine learning algorithms assume that there’s an infinite number of presents ahead, it makes more sense to quickly move forward and grab the presents because there are more ahead of you than behind.  The screenshot above is from a machine learning model playing Super Mario Kart, where it is easy to see how machine learning quickly gets better and better with successive plays of the game.

Closing out this episode, Michael and Michael include David Ma’s videos of food imagined in the style of  famous directors, a funny song about waffles and a discussion on Zork.  Hope you enjoy these tasty treats!

Selected links 

last week’s show, E177 — What could possibly go wrong? — http://gamesatwork.biz/2017/08/31/episode-177-what-could-possibly-go-wrong/

An open letter to the United Nations Convention on Certain Conventional Weapons — https://futureoflife.org/autonomous-weapons-open-letter-2017

We don’t have long to act — https://qz.com/1058280/we-do-not-have-long-to-act-teslas-tsla-elon-musk-and-others-warn-the-un-about-autonomous-weapons

Six Colors article:  Deep Learning Improves Siri’s voice — https://sixcolors.com/link/2017/08/deep-learning-improves-siris-voice/

Apple’s Machine Learning Journal:  Deep Learning for Siri’s Voice: On-device Deep Mixture Density Networks for Hybrid Unit Selection Synthesis — https://machinelearning.apple.com/2017/08/06/siri-voices.html

Uncanny Valley — https://en.wikipedia.org/wiki/Uncanny_valley

J.A.R.V.I.S. — https://en.wikipedia.org/wiki/Edwin_Jarvis#J.A.R.V.I.S.

O’Reilly article:  Bringing gaming to life with AI and deep learning — https://www.oreilly.com/ideas/bringing-gaming-to-life-with-ai-and-deep-learning

YouTube:  Super MarI/O Kart Commentary — https://www.youtube.com/watch?v=S9Y_I9vY8Qw

Food by Favorite Directors (Sort Of) — https://www.subtraction.com/2017/08/14/food-by-famous-directors-sort-of/

YouTube:  What if Alfonso Cuaron made pancakes?  — https://www.youtube.com/watch?v=6_hpJHNt4IE

Do You Like Waffles?  By Parry Gripp — https://itunes.apple.com/us/album/do-you-like-waffles/id468568946?i=468568983

MIT Technology Review — The Enduring Legacy of Zork — https://www.technologyreview.com/s/608670/the-enduring-legacy-of-zork/

MIT Technology Review article:  Siri for Business — https://www.technologyreview.com/s/600990/siri-for-business/

Play

Episode 60 – Bubbly Bubblers in Gamified Buildings

Happy Water CoolerPhaedra, Michael and Michael were delighted to be accompanied by Ross Smith, Director of Test in the Skype division of Microsoft for this show, centered around games for employee engagement.

Ross described the recent work that he and his team have done with giving personality to buildings and using the building itself as the ‘game master’, to orchestrate and facilitate employee engagement.

Games for health
One of the intriguing examples (also discussed in Episode 58, but without the benefit of Ross’s first hand experience) uses a water cooler with personality as the mechanism for gamifying health.  While drinking eight glasses of water a day is a pretty well known health tip (I’m drinking my 4th big glass as I’m writing this!), it is easily forgotten in the hectic business day.  The team at Microsoft gave their water cooler a bubbly user interface and 4 emotional modes (sad, happy, attract/flirt and party) to experiment with increasing the amount of water consumed by the building’s people.  The results:  more water for more people, and a wealth of data, feedback and ideas to continue the innovation.  The picture above was inspired by Ross’ story and is my low-tech example of giving the water cooler in my apartment some personality as well!
Another example is classifying the food choices in the cafeteria as green, yellow and red, and then using that data to guide behavior.  If less healthy food choices are selected, the building might suggest using the stairs or a longer route for a meeting later in the day.  This has helped the team learn more about what people are willing to share, and what they are not, and inform future game design on what should have individual vs team collective goals to maximize participation while respecting privacy.

Games for social / business interaction
The game ISHI was used as the buildings were being opened to help the team come together in an open office, connected community.  Ross described how people may respond to some survey questions indicating preference between skiing and snowboarding, or cat / dog fandom, and then how that kind of data can be used to create groups and teams with common interests and help people to get to know one another even better.  The building invited people to come to a central space, then you have met a number of people with common interest across job roles and functions, with the goal of beating the clock to figure out what do they have in common.  This method of dynamic team creation could be leveraged to bring a creative team together to brainstorm how to solve a client’s problem just as easily.
Windows Language Quality Game was another game with similar tremendous results, this game wound up attracting more than 4,000 people play and review over half a million screens, separating actionable feedback from non-actionable feedback received from public beta testers of Windows software.  Ross noted that rigorous data collection is key, and his experience of 4x – 16x increased participation and results because of the game mechanics can demonstrate even to the skeptics the business impact and win them over.

We were delighted to have Ross on the show with us today, and if you would like to learn more about him and the projects he’s involved in, you can find him on the following links:

Web www.42projects.org
Twitter   @42projects
Facebook  www.facebook.com/42projects
blog  http://42projects.blogspot.com/

We are happy to have had you dialed our way.  You can find us on G+, Tumblr, Facebook, Twitter and more.  Please comment, friend, tweet and send us topics you would like to hear more about!

Selected show links
ISHI game for the Skype team at Microsoft — http://research.microsoft.com/apps/video/default.aspx?id=200275
ISHI game documentary rough cut — http://www.youtube.com/watch?v=l0W431tP6SM&feature=youtu.be
Water Cooler — http://www.businessinsider.com/like-to-flirt-at-the-office-water-cooler-a-microsoft-intern-built-a-water-cooler-that-flirts-back-2013-8
Vote for Sentient Buildings for SXSW — http://panelpicker.sxsw.com/vote/21393
Microsoft Interns — http://research.microsoft.com/en-us/news/features/interns13-082613.aspx
Magic Circle of Play — http://en.wikipedia.org/wiki/Magic_Circle_(virtual_worlds)
Everymove.org — https://everymove.org
Chore Wars — http://www.chorewars.com
uTest — http://www.utest.com

Games we’re playing (or thinking about)
Madden 25 — http://www.easports.com/madden-nfl
Assassin’s Creed — http://assassinscreed.ubi.com/en-us/home/index.aspx
SimCity — http://www.simcity.com
Bioshock 2 — http://www.bioshock2game.com/en/
Call of Duty Black Ops 2 — http://www.callofduty.com/blackops/
iPad games for toddlers?  Phaedra would like your suggestions!
Dragon Box — http://research.microsoft.com/apps/video/default.aspx?id=200275
Motor World — https://itunes.apple.com/us/app/motor-world-car-factory/id580666714?mt=8
Plants vs Zombies 2 – https://itunes.apple.com/us/app/plants-vs.-zombies-2/id597986893?mt=8

Play

Episode 55 – Do Not Play This Game At Work

Tom Grant, analyst from Forrester Research, joins the show to talk about his new website Serious Games At Work and his love of board games. We look at a listener link on do Games at Work really help productivity. Realizing they are probably asking the wrong question. They should be asking – Should you be playing pirates games at work? If so, who get’s the red stapler?

Seriously, we discuss that the game part has to be more of a periodic entry into the Magic Circle, and not a mandatory daily activity. This allows businesses to achieve effective optimization, while still keeping the game fresh and engaging. Have you thought about refreshing your serious game content to keep your employees or customers engaged?

Show Links:
Serious Games At Work
Twitter
Crusader Kings II
Scourge of War – Gettysburg
Breakaway Games
Avalon Hill
World Boardgame Championships
Homo Ludens
Buy A Feature
Do Games At Work Help Productivity
Microsoft’s Language Quality Game

Obligatory causal reference to Daniel Pink

Monopoly removes jail
The Dinner Party Download
The Rant Episode

Saving Throws:
Emotiv Insight Kickstarter
Jaw Bone Up Team
Hospital Process Game
Serious Games At Work
Endgame Eurasia

Play