LESSWRONG
LW

All of Valerio's Comments + Replies

Towards a Less Bullshit Model of Semantics
Valerio2-3

I am skeptical about your theory of impact for investigating the question of which concepts would be convergent across minds, specifically your expectation that concepts validated through linguistic conventions may assist in non-ad-hoc interpretability of deep learning networks. Yet, I am interested in investigating semantics for the purpose of alignment. Let me try to explain how my model differs from yours.

First, for productively studying semantics, I recommend keeping a distinction between a semantics for vision (as the prototypical sensory input) and o... (read more)

Reply
My AI Model Delta Compared To Yudkowsky
Valerio10

As the internal ontology takes on any reflective aspects, parts of the representation that mix with facts about the AI's internals, I expect to find much larger differences

It could be worth exploring reflection in transparency-based AIs, the internals of which are observable. We can train a learning AI, which only learns concepts by grounding them on the AI's internals (consider the example of a language-based AI learning a representation linking saying words and its output procedure). Even if AI-learned concepts do not coincide with human concepts, becaus... (read more)

Reply
Views on when AGI comes and on strategy to reduce existential risk
Valerio20

I like your arguments on AGI timelines, but the last section of your post feels like you are reflecting on something I would call "civilization improvement" rather than on a 20+ years plan for AGI alignment. 

I am a bit confused by the way you are conflating "civilization improvement" with a strategy for alignment (when you discuss enhanced humans solving alignment, or discuss empathy in communicating a message "If you and people you know succeed at what you're trying to do, everyone will die").  Yes, given longer timelines, civilization improveme... (read more)

Reply
Where are the people building AGI in the non-dumb way?
Answer by Valerio-30

I am also interested in interpretable ML. I am developing artificial semiosis, a human-like AI training process which can achieve aligned (transparency-based, interpretability-based) cognition. You can find an example of the algorithms I am making here: the AI runs a non-deep-learning algorithm, does some reflection and forms a meaning for someone “saying” something, a meaning different from the usual meaning for humans, but perfectly interpretable.

I support then the case for differential technological development:

There are two counter-arguments to this th

... (read more)
Reply
The "semiosis reply" to the Chinese Room Argument
Valerio10

In his paper, Searle brings forward a lot of arguments.

Early in his argumentation and referring to the Chinese room, Searle makes this argument (which I ask you not to mix with later arguments without care):

it seems to me quite obvious in the example that I do not understand a word of the Chinese stories. I have inputs and outputs that are indistinguishable from those of the native Chinese speaker, and I can have any formal program you like, but I still understand nothing. For the same reasons, Schank's computer understands nothing of any stories.
... (read more)
Reply
3TruePath
Searle can be any X?? WTF? That's a bit confusingly written. The intuition Searle is pumping is that since he, as a component of the total system doesn't understand Chinese it seems counterintuitive to conclude that the whole system understands Chinese. When Searle says he is the system he is pointing to the fact that he is doing all the actual interpretation of instructions and is seems weird to think that the whole system has some extra experiences that let it understand Chinese even though he does not. When Searle uses the word understand he does not mean demonstrate the appropriate input output behavior he is presuming it has that behavior and asking about the system's experiences. Searle's view from his philosophy of language is that our understanding and mening is grounded in our experiences and what makes a person count as understanding (as opposed to merely dumbly parroting) Chinese is that they have certain kinds of experiences while manipulating the words. When Searle asserts the room doesn't understand Chinese he is asserting that it doesn't have the requisite experiences (because it's not having any experiences) that someone would need to have to count as understanding Chinese. Look, I've listened to Searle explain this himself multiple times during the 2 years of graduate seminars on philosophy of mind I took with him and have discussed this very argument with him at some length. I'm sorry but you are interpreting him incorrectly. I know I'm not making the confusion you suggest because I've personally talked with him at some length about his argument.
The "semiosis reply" to the Chinese Room Argument
Valerio10

Uhm, an Aboriginal tends to see meaning in anything. The more the regularities, the more meaning she will form. Semiosis is the dynamic process of interpreting these signs.

If you were put in a Chinese room with no other input than some incomprehensible scribbles you will probably start considering that what you are doing has indeed a meaning.

Of course, a less intelligent human in the room or a human put under pressure would not be able to understand Chinese even with the right algorithm. My point is that the right algorithm enables the right human to understand Chinese. Do you see that?

Reply
1binary_doge
Then that's an unnecessary assumption about Aboriginals. Take a native Madagascan instead (arbitrary choice of ethnicity) and he might not. As far as I know it is not true, and certainly not based on any concrete evidence, that humans must see intentional patterns in everything. Not every culture thought cloud patterns were a language for example. In such a culture, the one beholding the sky doesn't necessarily think it displays the actions of an intentful agent recording a message. The same can be true for Chinese scribbles. If what you're saying was true, it would be a very surprising fact that there are a whole bunch of human cultures in history that never invented writing. At any rate, if there exists a not-an-anomaly-example of a human that given sufficient time could not learn Chinese in a Chinese Room, the entire argument as a solution to the problem doesn't hold (lets call this "the normal man argument"). If it were enough that there exists a human that *could* learn Chinese in the room, then you could have just given some example of really intuitive learners throughout history or some such. It is enough for the original Chinese room to show a complete system that emulates understanding Chinese, but no part of it (specifically the human part) understands Chinese, and therefore you can't prove a machine is "actually thinking" and all that jazz because it might be constructed like the aforementioned system (this is the basis for the normal man argument). Of course, there are answers to this conundrum, but the one you posit doesn't contradict the original point.
The "semiosis reply" to the Chinese Room Argument
Valerio10

A more proper summary would read as follows:
1. P is an instantiated algorithm that behaves as if it [x]. (Where [x] = “understands and speaks Chinese”.)
2. If we examine P, we can easily see that its inner workings cannot possibly explain how it could [x].
3. Therefore, the fact that humans can [x] cannot be explainable by any algorithm.

I have some problem with your formulation. The fact that P does not understand [x] is nowhere in your formulation, not in premise #1. Conclusion #3 is wrong and should be written as "the fact that humans can [x] cannot ... (read more)

Reply
2Said Achmiz
Yes it is. Reread more closely, please. That is not Searle’s argument. I don’t think anything more may productively be said in this conversation as long as (as seems to be the case) you don’t understand what Searle was arguing.
The "semiosis reply" to the Chinese Room Argument
Valerio10

SCA infers that "somebody wrote that" where the term "somebody" is used more generally than in English.

SCA does not infer that another human being wrote that, but rather that a casual agent wrote that, maybe spirits of the caves.

If SCA enters two caves and observes natural patterns in cave A and the characters of "The adventures of Pinocchio" in cave B, she may deduce that two different spirits wrote them. Although she may discover some patterns in what spirit A (natural phenomena) wrote, she won't be able to discover a g... (read more)

Reply
1binary_doge
But the fact that it is purposeful writing, for example by a spirit, is an added assumption... SCA doesn't have to think that, she could think its randomly generated scribbles made by nature. Like how she doesn't think the rings on the inside of a tree are a form of telling a story. They are just meaningless signs. And if she does not think the signs have meaning, your statements don't follow (having scribbles doesn't mean that some other agent necessarily made them, and since the scribbles don't point to anything in reality there is no way to understand that P and p are of some same type of item). Thus, there exists a human to be put in a Chinese Room that can make the room replicate the understanding of Chinese without knowing Chinese herself.
The "semiosis reply" to the Chinese Room Argument
Valerio10

TruePath, you are mistaken, my argument addresses the main issue of explaining computer understanding (moreover, it seems that you are making confusion between the Chinese room argument and the “system reply” to it).
Let me clarify. I could write the Chinese room argument as the following deduction argument:
1) P is a computer program that does [x]
2) There is no computer program sufficient for explaining human understanding of [x]
=> 3) Computer program P does not understand [x]
In my view, assumption (2) is not demonstrated and the argument should be ... (read more)

Reply
1TruePath
If you want to argue against that piece of reasoning give it a different name because it's not the Chinese room argument. I took multiple graduate classes with professor Searle and, while there are a number of details Said definitely gets the overall outline correct and the argument you advanced is not his Chinese room argument. That doesn't mean we can't talk about your argument just don't insist it is Searle's Chinese room argument.
3Said Achmiz
This is not at all correct as a summary of Searle’s argument. A more proper summary would read as follows: 1. P is an instantiated algorithm that behaves as if it [x]. (Where [x] = “understands and speaks Chinese”.) 2. If we examine P, we can easily see that its inner workings cannot possibly explain how it could [x]. 3. Therefore, the fact that humans can [x] cannot be explainable by any algorithm. That the Room does not understand Chinese is not a conclusion of the argument. It’s taken as a premise; and the reader is induced to accede to taking it as a premise, on the basis of the “intuition pump” of the Room’s description (with the papers and so on). Now, you seem to disagree with this premise (#2). Fair enough; so do I. But then there’s nothing more to discuss. Searle’s argument collapses, and we’re done here. The rest of your argument seems aimed at shoring up the opposing intuition (unnecessary, but let’s go with it). However, it would not impress John Searle. He might say: very well, you propose to construct a computer program in a certain way, you propose to expose it to certain stimuli, yes, very good. Having done this, the resulting program would appear to understand Chinese. Would it still be some deterministic algorithm? Yes, of course; all computer programs are. Could you instantiate it in a Room-like structure, just like in the original thought experiment? Naturally. And so it would succumb to the same argument as the original Room.
Open thread, July 31 - August 6, 2017
Valerio00

Daniel, I'm curious too. What do you think about Fluid Construction Grammar? Can it be a good theory of language?

Reply
Steelmanning the Chinese Room Argument
Valerio00

cousin_it, aren't you forgetting that the rules of the Chinese Room are different than those of Turing's imitation game? While Turing does not let you in the other test room, Searle grants you complete access to the code of the program. If you could really work out a (Chinese) brain digital upload, you could develop a theory of consciousness/intelligence/intentionality from it. Unfortunately, artificial neural networks bear no connection to the brain, like ELIZA bears no connection to a human!

Reply