lunes, 28 de agosto de 2017

Podría ser una parte de la propia naturaleza de la inteligencia el hecho de que sólo una fracción está sujeta a explicaciones…

…resulta fácil imaginar que al recibir una recomendación de un restaurante, se quiera conocer el razonamiento subyacente.

Smart Lighting – El secreto más oscuro de la inteligencia artificial: ¿por qué hace lo que hace?

Adam Ferriss

La misteriosa mente de este vehículo señala un problema cada vez más palpable de la inteligencia artificial. La tecnología de IA del coche, conocida como aprendizaje profundo ha tenido mucho éxito resolviendo problemas en los últimos años, y cada vez se usa más en labores como generar subtítulos, reconocer la voz y traducir idiomas. Estas mismas técnicas podrían llegar a ser capaces de diagnosticar enfermedades mortales, tomar decisiones bursátiles multimillonarias y transformar industrias al completo.

Pero esto no sucederá  (o no debería) a menos que consigamos que las técnicas como el aprendizaje profundo resulten más comprensibles para sus creadores y rindan cuentas ante los usuarios. En caso contrario, resultará difícil predecir cuándo se podrían producir fallos, los cuales son inevitables. Este es uno de los motivos por los que el coche de Nvidia aún es experimental.

Ya se están empleando modelos matemáticos para ayudar a determinar quién recibe la libertad condicional, quién es apto para obtener préstamos y quién es contratado para ocupar un puesto vacante. Si se pudiese acceder a estos modelos matemáticos, sería posible entender su razonamiento. Pero los bancos, el ejército y los empleadores están centrándose en enfoques de aprendizaje automático aún más complejos que podrían resultar totalmente inescrutables.



El líder del equipo de Siri de Apple, Tom Gruber, dice que el carácter explicable es una consideración clave para su equipo en sus esfuerzos por hacerla más inteligente y hábil. Gruber no quiso hablar de planes específicos para el futuro de Siri, pero resulta fácil imaginar que al recibir una recomendación de un restaurante, se quiera conocer el razonamiento subyacente. El director de investigaciones de IA de Apple y profesor de la Universidad de Carnegie Mellon (EEUU), Ruslan Salakhutdinov, considera que la capacidad de explicarse es el núcleo de la relación en evolución entre los humanos y las máquinas inteligentes. “Va a generar confianza”, afirma.



Al igual que muchos aspectos del comportamiento humano resultan imposibles de explicar en detalle, tal vez no será posible que la inteligencia artificial llegue a explicar todo lo que hace. “Incluso si alguien te puede dar una explicación razonable [de sus acciones], probablemente estará incompleta, y lo mismo podría aplicarse a la inteligencia artificial”, sugiere Clune. El experto añade: “Simplemente podría ser una parte de la propia naturaleza de la inteligencia el hecho de que sólo una fracción está sujeta a explicaciones racionales. Parte de ella es simplemente instintiva, o subconsciente o inescrutable”.



Al igual que la sociedad se construye sobre una base de comportamientos aceptables, necesitaremos diseñar los sistemas de IA para respetar y encajar con nuestras normas sociales. Si vamos a crear tanques robóticos y otras máquinas de matar, es importante que su toma de decisiones concuerde con nuestros juicios éticos.

Daniel Dennett, que estudia la consciencia y la mente. Un capítulo de su último libro, From Bacteria to Bach and Back, un tratado enciclopédico sobre la consciencia, sugiere que una parte natural de la evolución de la propia inteligencia consiste en desarrollar sistemas capaces de ejecutar tareas que sus creadores son incapaces de ejecutar. “La pregunta es: ¿qué adaptaciones tenemos que hacer para hacerlo bien, qué estándares debemos exigirles que cumplan a ellos, y a nosotros mismos?”, me pregunta dentro de su abarrotado despacho en el idílico campus de la universidad.

También me hizo una advertencia sobre la búsqueda del carácter explicable: “Creo que si vamos a utilizar estas cosas y depender de ellas, entonces necesitamos el mejor entendimiento posible de cómo y por qué nos proporcionan respuestas”, sugiere. Pero, puesto que podría no existir ninguna respuesta perfecta, deberíamos mostrarnos igual de cautelosos ante las explicaciones de la IA como nos mostramos ante las explicaciones humanas, independientemente de lo lista que parezca la máquina. Dennet concluye: “Si no puede explicar lo que hace mejor que nosotros, entonces no te fíes”.

sábado, 26 de agosto de 2017

These 7 Forces Are Changing the World at an Extraordinary Rate

by @singularityhub based on @PeterDiamandis talk(s).
if a genius was born in a remote village 100 years ago, she would likely not have been able to gain access to the resources needed to put its gifts to widely productive use
These 7 Forces Are Changing the World at an Extraordinary Rate

  • Computation
  • Convergence
  • Interface Moments
  • Connectivity
  • Sensors
  • Intelligence
  • Wealth Concentration


domingo, 20 de agosto de 2017

5 ways to get out of your own way and communicate clearly

by @saralindberg13
I still find some trouble listening to the intent behind the words and not making assumptions.

Headspace blog – 5 ways to get out of your own way and communicate clearly B
Be impeccable with your word.Avoid taking things personally.Avoid making assumptions. The problem with making assumptions is that they’re really only a version of our own observations and feelings.Always do your best.Be skeptical, but learn to listen. “Listen to the intent behind the words, and you will understand the real message.”


sábado, 19 de agosto de 2017

Designing For a Socially Valuable User Experience

@nicolashenchoz from @epflecallab Designing For a Socially Valuable User Experience - Billionaire
“Design,” he says, “is not primarily about artistic form, but about a socially valuable user experience. One that combines cultural, functional and emotional dimensions.”
“We don’t strive towards futuristic forms that sometimes can appear anecdotal and gimmicky, but for the sense of normality that makes you feel comfortable enough to adopt. When we work on augmented reality projects, we add digital information around a physical object in the most natural way, to value the content.” He adds: “The ethics of design are to be clear with your intention in what you design. We want something new to be relevant. Meaning has a longer lifecycle than technology.”
When asked to describe the perfect design project, Henchoz offers up a surprising response that indicates his commitment to due process: “Design research. It is critical that design should begin to produce knowledge that tracks the impact of an artefact, creating a feedback loop as in scientific methodology.” The EPFL+ECAL Lab employs psychologists to analyse qualitative and quantitative data at consumer level and track the effect on the user, including emotional responses. “Sustainable innovation relies on useful paradigms — not a six-month thrill of the new,” says Henchoz.

Designing For a Socially Valuable User Experience

@nicolashenchoz from @epflecallab Designing For a Socially Valuable User Experience - Billionaire
“Design,” he says, “is not primarily about artistic form, but about a socially valuable user experience. One that combines cultural, functional and emotional dimensions.”
“We don’t strive towards futuristic forms that sometimes can appear anecdotal and gimmicky, but for the sense of normality that makes you feel comfortable enough to adopt. When we work on augmented reality projects, we add digital information around a physical object in the most natural way, to value the content.” He adds: “The ethics of design are to be clear with your intention in what you design. We want something new to be relevant. Meaning has a longer lifecycle than technology.”
When asked to describe the perfect design project, Henchoz offers up a surprising response that indicates his commitment to due process: “Design research. It is critical that design should begin to produce knowledge that tracks the impact of an artefact, creating a feedback loop as in scientific methodology.” The EPFL+ECAL Lab employs psychologists to analyse qualitative and quantitative data at consumer level and track the effect on the user, including emotional responses. “Sustainable innovation relies on useful paradigms — not a six-month thrill of the new,” says Henchoz.

domingo, 13 de agosto de 2017

24 Industries Other Than Auto Driverless Cars Could Turn Upside Down

24 Industries Other Than Auto Driverless Cars Could Turn Upside Down

1. INSURANCE
2. AUTO REPAIRS
3. PROFESSIONAL DRIVERS AND TRUCKING
4. HOTELS
5. AIRLINES
6. AUTO PARTS
7. RIDE-HAILING
8. PUBLIC TRANSPORTATION
9. PARKING GARAGES AND LOTS
10. FAST FOOD
11. ENERGY AND PETROLEUM
12. REAL ESTATE
13. MEDIA AND ENTERTAINMENT
14. DELIVERIES
15. BRICK AND MORTAR RETAIL
16. AUTO DEALERSHIPS
17. OIL-CHANGE SHOPS AND CAR WASHES
18. HEALTHCARE
19. DRIVING SCHOOLS
20. URBAN PLANNING
21. INTERNET SERVICE PROVISION
22. ‘INTERIOR’ DESIGN/MANUFACTURING
23. CYBERSECURITY
24. TRAFFIC ENFORCEMENT

End of Typing

An Open Letter To CEOs

by @AlexOsterwalder & @ypigneur Strategyzer - An Open Letter To CEOs
You have been excellent at executing and improving your proven and successful business models. But as the research above shows, you have not yet found the answer to inventing entirely new business models, value propositions, and growth engines. 
In fact, managing the present is taking oxygen away from inventing the future.

You not only have to be world class at executing and improving your current business model, but you also have to be world class at searching and inventing new business models for the future. 
That’s the real leadership challenge.

Your innovation engine is not a space where you write business plans for new ideas. Your main goal is to decrease the risk and uncertainty around new ideas. It’s a space where you prototype and test new business models and value propositions; where you experiment and gather evidence as cheaply and quickly as possible by getting out the building with methodologies like Lean Startup and Customer Discovery.

On one hand your execution engine will need to be world class at managing factories and tolerating zero failure; and on the other hand, your innovation engine will need to be world class at experimenting, failing, and learning to shape new ideas. 
Lastly, your innovation engine will need help from your execution engine--we cannot stress this enough.

How to Take a Full-Page Screenshot (with Chrome)

by @zapier How to Take a Full-Page Screenshot
Start by using the shortcut pairs below—enter the first shortcut, followed by the second—depending on your operating system:
On Mac1. Alt + Command + I2. Command + Shift + POn Windows/Linux/Chrome OS1. Ctrl + Shift + I2. Ctrl + Shift + PThese keyboard shortcuts will open Chrome's developer menu. Just type "screenshot" and you'll see the option appear to "capture full size screenshot." Simply select this and Chrome will automatically save a full-page screenshot to your Downloads folder!


miércoles, 9 de agosto de 2017

¿La brecha es económica y no digital?

De alguna manera he visto estos tres artículos muy interrelacionados.  ¯\_()_/¯

1.
Un interesante artículo, en el que parece defenderse que la gente se independice, beba y fume (y se ponga a trabajar pronto para podérselo pagar) como en 1970 ó 1980… en lugar de estar en la habitación mirando la pantalla de su smartphone. ;-P
Porque claro la tele no tiene pantalla, y los Commodore (como la tele) no son dispositivos electrónicos.
Have Smartphones Destroyed a Generation? - The Atlantic
But the allure of independence, so powerful to previous generations, holds less sway over today’s teens, who are less likely to leave the house without their parents. The shift is stunning: 12th-graders in 2015 were going out less often than eighth-gradersdid as recently as 2009.
A lo mejor unos padres centrados en objetivos "por el bien de sus hijos" muy distintos a los que se fijaban padres de generaciones anteriores son los que han traído a las redes sociales a aprovechar todo ésto. No es cosa del smartphone.

Que las redes sociales pueden hacerte infeliz… seguro; sobre todo cuando tu educación se centra en tus resultados, tus títulos, la cantidad de actividades extracurriculares, lo que pareces… para que los padres se puedan sentir orgullosos. Y sí, ese es el juego de las redes sociales y esa tan profesional en la que estamos los padres, LinkedIn, también va de lo mismo que las demás.

*****
Otros dos artículos me han parecido también muy interesantes y reveladores sobre la coincidencia con el desarrollo del smartphone y la nueva realidad económica derivada de la crisis financiera de 2007/2008 y ese acrecentamiento de la brecha entre los que más y los que menos tienen.

2.
El primero, en Wall Street Journal (de pago, deberías podéis leerlo aquí), habla de cómo las compañías quieren ganar al próximo "billón" de usuarios de smartphones, a través de conseguir a bajo coste de hardware aplicaciones que requieran poca alfabetización, vídeo y voz en mercados emergentes…

Instead of typing searches and emails, a wave of newcomers—“the next billion,” the tech industry calls them—is avoiding text, using voice activation and communicating with images. They are a swath of the world’s less-educated, online for the first time thanks to low-end smartphones, cheap data plans and intuitive apps that let them navigate despite poor literacy.
Incumbent tech companies are finding they must rethink their products for these newcomers and face local competitors that have been quicker to figure them out. “We are seeing a new kind of internet user,” said Caesar Sengupta, who heads a group at Alphabet Inc.’s Google trying to adapt to the new wave. “The new users are very different from the first billion.”




3.
El segundo, es una muestra de como no queremos perder esa alfabetización, habrá que seguir sabiendo leer y escribir A MANO, que parece que así es más fácil aprender

It may well be that the physicality of shaping letters cements concepts in the mind. For example, to type the word “typing,” I made the same motion on the keyboard six times, choosing which letter to type but not forming them. But if I were to write the same thing by hand, I’d have to shape six different letters and put them together. That takes more effort and seems to both demand more of the brain and leave a deeper imprint on the mind than typing. That imprint appears to be critical when learning new things.

miércoles, 2 de agosto de 2017

How To Spot and Spark Flow

by @uicynthia Track and Facilitate Your Engineers’ Flow States In This Simple Way | First Round Review

‘The more a job inherently resembles a game — with variety, appropriate and flexible challenges, clear goals, and immediate feedback — the more enjoyable it will be regardless of the worker’s level of development.’ ~ Csikszentmihalyi


Perhaps more important than recognizing flow, then, is recognizing its absence. Typically, engineers who fall out of flow will land in one of three common ruts:
Apathy — Low Skill and Low Challenge
Maxwell looks for one big red flag to spot this one: a consistent failure to chime into the conversation. An apathetic team member offers no suggestions about product features or team processes. They don’t share an opinion of job candidates, and try to avoid weighing in on ideas when asked. “It's almost as though they're hoping they don't get called on,” she says. “They're not being challenged, and they don't care.”
Managers at larger companies in particular may need to look out for one specific type of apathetic employee: “Rester Vesters” or people who just phone it in until their stock options fully vest.

Anxiety — Low Skill and High Challenge

When someone is pushed to tackle challenges their skill set can’t accommodate — or at a pace that isn’t feasible — anxiety is an understandable outcome. Maxwell urges leaders to keep an ear out for telltale anxiety-based phrases, and know how to translate them.
“Listen for things like, ‘Oh, this is a speculative fix,’ or ‘I wouldn't normally do things this way, but given the time constraints…,’” she says. “People who are feeling anxiety might also start blaming others for not meeting a deadline. Or they may say, ‘I have too much on my plate right now.’"

Boredom — High Skill and Low Challenge
“Usually people will fall into boredom because their skillset has increased. They've taken a leap forward, they’ve learned a lot. Maybe they just shipped something or conquered an obstacle, and now they don't feel like they're being challenged,” says Maxwell.
A bored engineer is usually executing the same tasks again and again, finishing them quickly and then spinning their wheels waiting for a new assignment. “Boredom might manifest as resentment over how projects are being assigned across the team,” says Maxwell. Keep an eye out, too, for anyone who creates unnecessary projects, or over-engineers simple problems, just to flex underused muscles. “If you see an increase in exotic, latest-and-greatest libraries entering your codebase, you might have a bored team.”

Between The Extremes

Of course, human emotion — like human workplaces — is complex, and people will likely spend some portion of their time moving between states. Those periods of transition, as someone moves out of flow in one direction or another, can be particularly impactful times to guide employees back toward flow:

Doubt — From Flow Toward Anxiety
When someone takes on greater challenges, without expanding their skill set to meet them, it’s logical that they would begin to experience self-doubt. In these cases the person suspects they’re lacking the skills to achieve the task. They might start to take offense to otherwise harmless code reviews. Or get stuck in analysis paralysis. They might spend a lot of time seeking advice from senior developers.
Left unchecked, the danger is that doubt becomes infectious. Engineers in this state may begin to not only doubt the value of the project but their leadership team, too. “They may begin to think, ‘Are they making good decisions? Have they put us in an unreasonable timeline? Are they asking us to do something that's just not possible?’" says Maxwell.
When you see reports plotting themselves on the graph in this direction, encourage them to speak up about any concerns they’re having. Simply acknowledging those feelings of doubt can be the fastest way to move back into flow. “One good rule of thumb is think smaller. That means breaking a task into smaller parts to build up confidence. Or celebrating smaller wins than you would otherwise,” says Maxwell. “If that fails, they may need some scaffolding. Get the person in a pair-programming situation so the world doesn’t feel like it’s resting only on their shoulders. As the person regains confidence, introduce more independent work.”

Nostalgia — From Flow Toward Boredom
The move toward boredom typically follows a period of skill-building; the engineer no longer feels challenged because they’ve grown. Falling out of flow in that direction is often marked by feelings of nostalgia. “They want to recreate what it felt like to be in flow, and they don't know exactly how,” says Maxwell.
There’s a hopefulness to this moment, though. A person experiencing nostalgia doesn’t want to be bored; they want to recapture the feeling of learning and growing. And that’s prime material for a manager looking to coax that engineer back into a productive state. “Help them identify what they’re nostalgic for. For example, is it the size of the team, or the ambition of the project?” Work with them to recreate the conditions they’re missing, while they’re still fresh in their mind.
The best managers will use every tool at their disposal to understand what’s going on within their teams, including, of course, their own observations. But while looks are famously deceiving, quantitative assessments are more concrete. In many cases, giving your team an objective way to articulate dissatisfaction is the only way to uncover important trends.

martes, 1 de agosto de 2017

The hard part is getting it (the data) organized and figuring out what’s relevant to your process.

Robo-Advisors Aren't A Difficult Tech Problem And Are Already On Their Way To Being Commoditized

Robin: Well, the buzzwords, I’ve noticed, are very much in the financial industry… Is machine learning. But, basically, using more modern artificial intelligence techniques in everything from investing to solving problems that they have on the consumer side, [inaudible 01:01:00]
How much of this is, to return to the cool kids trying to sound with the times, [inaudible 01:01:05] machine learning is very much [inaudible 01:01:07], but do you think, is this a future we’re gonna see more things move that way?
Joe: Yeah. That one, I’m more skeptical of a lot of the efforts I’ve seen in Wall Street. So it might be a little bit more of the cool kids syndrome. I’ve had several CIOs of big institutions ping me saying they have a big machine learning project, and can I connect them to the right people for what they are doing recently.
I mean, machine learning is a tool that all the great technology companies are using now to get done certain things. When you have a bunch of data and you need to figure out what the best answer is, I think you apply it.
I mean, it’s kind of funny, right? It’s artificial intelligence, and once it works we call it machine learning, and there’s all this research on the AI front. I’m definitely bullish on what AI is making possible, but I don’t think it’s this panache or this thing that solves everything. I think the harder problem, frankly, is getting all of your data organized in an infrastructure where you could use it for the problems.
So I think the hard thing to say is, what data should we be using to solve this problem? What is that process? Then once you have all of the data organized, you can hire a single Ph.D. from Stanford. He will do your machine learning thing for you. That’s not that hard. The hard part is getting it organized and figuring out that it’s relevant to this process.
So that’s much more of a strategy and process question to me.
Robin: But you think that it’s gonna be more, I mean, on artificial intelligence. I like that definition, that artificial intelligence is everything we haven’t been able to do that. Everything else is [inaudible 01:02:21]
But how much will we be able to do in the future? I mean, looking forward a little bit.
Joe: Sure.
So I definitely have a few investments in artificial intelligence companies that are trying to push the boundary and change things. I mean, I was actually really surprised that Demis and the team at DeepMind were able to solve the Go problem.
I think that was pretty impressive. But even looking into it, it’s not really capturing, in my view, human creativity, or anything even close to that yet. So I think we’re still decades and decades away from needing to worry about being replaced.
I think the much more interesting problem that we’re focusing on kind of the next 10 or 20 years is, how do we organize the infrastructure of our companies to get the data structured and available to use it for decisions?
I think it’s not gonna be the people who are best at AI and machine learning who win. It’s gonna be the people who are best at that data infrastructure and process problem. I mean, most of the top companies in Silicon Valley, like linear regressions and things, get you 90% of the way. I’m not seeing a lot of really, really hard AI that makes the big difference, for at least the things I’m seeing.
https://youtu.be/_lz2vCOxjOo