Transdisciplinarian (stats, datasci, ml, lang/socSci, tech, art, science, philosophy). (Use-inspired) fundamental research.Opinions my own. Accidental activist.
Zurich, Switzerland
Joined December 2021
- Tweets16.4K
- Following1.7K
- Followers295
- Likes776
Pinned Tweet
Education (or research) is NOT therapy or tutoring for professors at the expense of students' learning/career either.
@Bundesrat_CH @WhiteHouse @RoyalFamily @UN @EU_Commission @EFTAsecretariat @UNESCO and everyone else (including all nations or administrative regions and 1/
This ⤵️. "L"* and the pseudo-sciences thereof (in the name of "linguistics" or via cognitive science(s)) ought to be banned and these "professionals" arrested.
.@schmidtsciences .@NSF .@FBI .@CIA .@UN .@INTERPOL_HQ .@USSupremeCourt
*that which used to be referred to as "language"
We're excited to be able to continue our work (with @begusgasper) on rethinking language, thanks to @schmidtsciences HAVI.
We will test what properties of human language (sound change, dialect, and more) appear in nonhuman communication. This research lays the groundwork for reconceptualizing language as an evolutionary information technology applicable across human, animal, and machine communication.
.@UN
i. There is a right to freedom of expression and an ethical expectation and/or requirement to respect such (that makes communication).
ii. The concept of "nation"/"state" is independent of that of "l"*.
iii. Even if one were to go by an earlier misperception/malpractice 1/
In every language, everywhere, human rights belong to all of us.
They are non-negotiable.
Learn about the Universal Declaration of Human Rights: ohchr.org/en/universal-decla…
— via @UNHumanRights
this? xitter.cf/adawan919/status/21126…
If this is about human rights, then drop the use of "l"*. Human rights belong to all, regardless of "l"* or how one expresses oneself.
v. Design suggestions/ideas:
a. if every nation/state in the .@UN has a flag, use that instead. No "l"*;
3/
Whoever posted this did/does NOT understand why there is / should be no "l"*. Don't just chime along. THINK!
That, or it's blatant collusion.
Very disappointing. Whose idea is this? .@antonioguterres, do you know?
*that which used to be referred to as "language"
b. think: would "HUMAN RIGHS: our common 'race'/'colour'" be appropriate? (The answer is NO.)
xitter.cf/adawan919/status/21127…
*that which used to be referred to as "language" 4/4
(un.org/en/about-us/universal…)
Just like "race" and "colour" are un-/pseudo-scientific misconceptions that one should not be discriminated by, so is "language". (The "l" term, if one must, can/should only be used in a prohibition, to be outlawed.) 2/2
i. There is no need to "re-market" ML. ML can stay as is. There is no need for "AI".
ii. Ask "AI" practitioners what they get from my findings.
iii. Ask them the questions below---see what answers, if any, one gets. Then one'd know re-assessment & no "AI" is a necessary move.
What is even "SOTA"? Do the number of parameters matter? If/When so, what are the assumptions?
xitter.cf/adawan919/status/21126…
There can be tech, there can be ML. But no "AI".
Please share this message & be notified of my findings from 2019 on (incl. my posts/replies from social media).
And, as a reminder, when it comes world knowledge and values, the "small data"-interpretation is still rather dominant.
(Why my findings concerning that which used to be referred to as "language" matter.)
Is / Should the Transformer (be) considered SOTA?
For my SOTA modeling results, yes. It has been very helpful.
Do my findings concerning CNN (convolutional neural network, as shown under "'incomplete' experimental results" on my website sites.google.com/view/adawan), showing a 1/
Is / Should the Transformer (be) considered SOTA?
For my SOTA modeling results, yes. It has been very helpful.
Do my findings concerning CNN (convolutional neural network, as shown under "'incomplete' experimental results" on my website sites.google.com/view/adawan), showing a 1/
as supporter and/or user of either/any algorithm, being targeted by my findings? By me, NO. I had no intent of such kind. On my end, there is nothing personal.
xitter.cf/adawan919/status/21126…
An algorithm is just an algorithm. If one algorithm doesn't work as optimally, simply design 5/
This⤵️. Linguistics is fraud (& wrong) too.
If I hadn't been hurt, hadn't been kept busy with all the roadblocks, bureaucratic hurdles after my evaluative study that there is & shouldn't be any "l"*, I might've thought differently.
*that which used to be referred to as "language"
I thought these people would all be in jail by now. Why are they running around still?
Thank you!!! But I certainly can’t take credit for the whole group. So much of it due to all the students and the other faculty:
@jurafsky @percyliang @ChrisGPotts @YejinChoinka @MonicaSLam @Diyi_Yang @tatsu_hashimoto