edejong

Could a researcher chime in: is it possible that smaller models have a higher potential for mapping activations to abstractions that are present in our own language understanding and if so, could we use these makings to make our models more transparent and explainable?

Solution in Scala, with JaCoP:

  import org.jacop.scala._

  object Main extends App with jacop {
    def forAll(l: Seq[BoolVar]) = l.foldLeft[BoolVar](true) { case (r, m) => r /\ m}
    def forAny(l: Seq[BoolVar]) = l.foldLeft[BoolVar](false) { case (r, m) => r \/ m}
  
    val v = List.tabulate(6)(i => new BoolVar(s"Answer ${i + 1}"))
    v(0) #= forAll(v.drop(1))
    v(1) #= ~forAny(v.drop(2))
    v(2) #= forAll(v.take(2))
    v(3) #= forAny(v.take(3))
    v(4) #= ~forAny(v.take(4))
    v(5) #= ~forAny(v.take(5))
    satisfy(search(v, input_order, indomain_min))
    print("Solution: " + v.filter(_.domain.contains(1)).map(_.id).mkString(" "))
  }

Results in:

  Solution: Answer 5

Part of maximizing profit is dependent upon being able to attract and retain top performing human resources. Another part is public perception: brand value and customer engagement. Yet another part in dependent on their network of partners, who might want to refrain from doing business with parties engaged in unsound (unethical) business practices.

Since large parts of these are hidden variables and only come to light when certain actions are taken, I would say it is actually in the shareholders best interest for employees to complain and strike. The alternative is that top talent is leaving the company, partners untie relations and customers seek alternatives.

It's all about perception and communication.

Oh, the shame. I've handed in my resignation as grammar nazi.

And of grammar nazi’s telling you it’s “it’s” not “its”.

Unless it is a semantic edit.

I personally hate these very polluting moped abominations on the Dutch bicycle paths. They should’ve been banned (or heavily taxed) a decade ago when the Greens asked questions in parliament due to appalling emissions findings of a Dutch research institute. But, the coward politicians pointed to the EU and implicitly said the health of our bicyclists need to be decided in Brussels.

Neither to me. Perhaps the elderly, mopeds and weekends might bring the average down, but still. We regularly go on long bicycle trips of +30km. Some of my coworkers regularly race +80km.

disclosure: Dutch myself

Well, that's the funny thing.

The claim is: "No IP address == unhackable."

So, for a second stop assuming we are in a nuclear bunker. There are many unconnected systems that are still hackable. For example, the Stuxnet worm was able to infiltrate an unconnected SCADA system using USB exploits. Many systems radiate data in EM and ultrasonic (side-channel). And obviously, there are known and unknown backdoors in PCBs, ECs, and ICs. All of these do not require an IP address.

I would even argue that having an IP address can improve security by increasing monitoring, improving continuous security updates and decreasing a false sense of security. This was the original goal of DARPA all along.

> You can't hack something that doesn't have an IP address.

That's a very risky assumption to make. There are many attack vectors besides TCP/IP: social engineering, side-channel attacks, EM pulse, backdoors in ICs...

I came to the comments just to voice a similar concern. We need to take ownership of our data, whether it is biometric, environmental or social. And once we have established this for ourselves, we should make it as accessible as the commercial offerings. This should be the new open source movement. Projects like OpenHAB and homebrew sensor networks with low power open source schematics and drivers.

Hey, we are already heading there. It’s just the masses that will stay ignorant for a decade.

Do you use a search engine? Social media? Any modern online game? A connected smart home? A smart phone without its privacy settings altered? A modern flat screen tv? A modern car? All of them do behavioral extraction in one form of another. Often under the guise of entertainment.

To single out Ingress or Pokémon Go seems to ignore the much bigger problem.

And, as an Ingress oldie, it was a lot of fun. We planned for weeks to cover the northern hemisphere, preparing with people in four different continents. I’ve met people and seen parts of the world I would never have met or seen otherwise.

Love that illustration. Do you have a source? I'd like to pay attribution when I share it elsewhere.

By correlating on a shared code. But please don't take my word for it, I don't even have a license. See this [1] excellent Reddit discussion.

[1] https://www.reddit.com/r/amateurradio/comments/49yuwx/til_gp...

Time to unpack those radios that operate slightly under the noise floor, using shared correlation codes and synchronized clocks. P2P for the win.

[edit: grammar]

> Who is disputing that? I am 99.9% certain that this is true.

Well, the general horror story that is being told is that intelligence and the ability to use it go hand in hand. As soon as machines surpass our intelligence, all hell breaks loose and there is no turning back. The focus is very much on the individual (both human and machine) and not on the structure.

> They have to have the same motivation, interests, emotions and needs like us.

Here is where I need some more convincing. In my view, nature is full of systems and groups of species living in symbiosis. Due to the slow development of the thought framework, we will co-evolve with the machines, or become part machine ourselves. In a certain way, we already are.

Perhaps I don't really have the word for it and I am sorry for that. Our understanding of intelligence, whether mechanistic or biological, is mostly as points on a linear scale (IQ). I think this misses the big picture: we are not biological systems, but have lived in symbiosis with our child teknos since we made our first tools. In this regard, there is no subject nor object: they are one. I am fundamentally with Heidegger on this one.

Thank you so much. This comment is golden. I’ve been pushing similar thoughts for years. Also, we should take into consideration the power of our communication structure which has evolved over hundreds of generations.

Even if there would be true generalized Artificial intelligence, it is not unreasonable to think that the formation of the right thought framework for those systems to function effectively will require a substantial amount of time. It is also reasonable to think that such change would probably be in symbiosis with human intelligence. This change would not be sudden, but happen over many generations. It is my opinion that this change has already started at least a generation ago.

Perhaps you want to rebase your changes to the root of your Git repo (in other words, remove history). Now people can just check out the HEAD^ commit.

I didn't read the article but skimmed. Guilty as charged <blush-emoji> Thought it was about RPA (Robotic Process Automation).

Also, the 'clang' robots that come to mind have a certain amount of self-reliance. The 'non-clang' robots (which are basically macro's running in VMs), stop working as soon as an unexpected popup appears, the software is upgraded or the products on one page are suddenly on two.

No, the 'non-clang' robots are an indication of the software engineers' gross failure to properly integrate and standardize data exchange between solutions.

There’s certainly no physical force applied, but using the right techniques, we can predictively force a large part of the population to work against their own best interests using psychology alone. Obviously, this is an amalgam of social elements, redefined value system, stress response and sunk cost fallacy.

Borders to slavery in my books. It doesn’t even matter whether these are productive hours, it’s just forced labor using social anxiety peer pressure and FOMO to force people into doing your bidding.

Thought we had labor laws for that...

These are not my words. All I am arguing is that we should expect the simplest possible connection between nutrition and health. Although I do agree we should also understand the systematic effects of certain choices, we cannot make decisions if all these effects are aggregated haphazardly.

It would make sense if another study would combine the nutritional long-term effects, the societal costs, the availability of alternative sources of amino acids and other factors. Nevertheless, the conclusion: "Red meat consumption is unhealthy for individuals" should (imho) not consider the environmental effects. Principle of least astonishment.

> Consumers don't want to spend more for appliances. They want cheap.

Many consumers are smart enough to understand that 'cheap' is a property calculated over the lifetime of the product.

Harvard Response: "The panel declared “considerations of environmental impact” out of the scope of their recommendations. This is a missed opportunity because climate change and environmental degradation have serious effects on human health, and thus is important to consider when making recommendations on diet, even if this is addressed separately from direct effects on individual health."

My response: should we apply this to other topics as well? Pimp the efficiency of renewables compared with carbon-based? Up the effectivity of hydrogen-based energy distribution? Lift the results of a study promoting working from home?

Two words come to mind: Occam's razor.

Does ClassDojo provide supporting material with a didactic foundation on how to effectively use the app? Do schools or teacher associations provide such material? IMHO, without such guidance, it is irresponsible to use these apps.

In other words: the value is in the extraction of human behaviour which previously existed in the private sphere.

Our Samsung Flip is really snappy and relatively affordable. Still enjoy whiteboards more though.

People don’t make purely rational decisions, especially for ‘saviors’ and especially when it’s about respect.

You forget that she can be both a figurehead and a leader. That’s actually where the real problem is: as a ‘savior’ she anthropomorphises the climate crisis. Most people don’t know how to attack abstract ideas, but as soon as the idea gets a body, the party begins. IMHO, Thunberg could, in the long term, be a net negative to the climate discussion, because it allows for much stronger polarization.