Arnav Das's Avatar

Arnav Das

@arnav7.bsky.social

I try to "Observe, Reflect and Explore around" in different ways My posts here will be a mishmash of personal opinions and smol curious excursions..

1,784 Followers  |  12,315 Following  |  843 Posts  |  Joined: 30.07.2023  |  1.9469

Latest posts by arnav7.bsky.social on Bluesky

Sometimes I meet new people and immediately decide "we should not be friends" and this is okay. You do not have to be friends with everyone.

That said, it can be good to sometimes step out of your comfort zone, and relationship-building takes effort. But recognize that not everyone is compatible.

16.02.2026 17:46 β€” πŸ‘ 6    πŸ” 1    πŸ’¬ 1    πŸ“Œ 0
Preview
WA Democrats consider retreat on estate tax, fearing wealth exodus Democrats in the state Legislature are moving to roll back a big increase in the estate tax rate, acknowledging it may be causing some very wealthy people to move.

You can't call every step of creeping fascism a distraction from kitchen-table issues and then punt whenever you have a chance to address the kitchen-table issues.

16.02.2026 18:06 β€” πŸ‘ 1726    πŸ” 306    πŸ’¬ 52    πŸ“Œ 29
Preview
If progress is not to falter, students must be trained in open research The how and why of conducting transparent, rigorous, ethical research must be explicitly taught, say Madeleine Pownall, Charlotte Pennington and Flavio Azevedo

β€œOpen research is about more than the tightening of analytical and methodological standards. The movement also invites us to reconsider how, and by whom, knowledge is created, shared and evaluated”

By @maddipow.bsky.social, @drcpennington.bsky.social, & @flavioazevedo.bsky.social

#MetaSci #OpenSci

16.02.2026 18:10 β€” πŸ‘ 19    πŸ” 9    πŸ’¬ 0    πŸ“Œ 0
Preview
Metascience for whom? A question as old as science. Before we fix science, we need to ask who built it!

β€œWe should be careful not to marginalise questions of power, because in wearing that aura of clean objectivity, metascience risks becoming strangely depoliticised.”

By @batoolmm.bsky.social

16.02.2026 11:09 β€” πŸ‘ 22    πŸ” 5    πŸ’¬ 2    πŸ“Œ 2

I’m honored to be selected as a Humanities and AI Virtual Institute (HAVI) development awardee, with seed funding from @schmidtsciences.bsky.social, as a co-PI on two of eleven interdisciplinary teams exploring AI, human history, and culture.

1/2

16.02.2026 14:52 β€” πŸ‘ 63    πŸ” 3    πŸ’¬ 3    πŸ“Œ 1

Pure moral panic. How, precisely, would a child who experiments with a different haircut or pronouns get β€œlocked in” to a trans identity?

The parents who had no problem letting her identify as trans suddenly restrict her from identifying as cis? It makes no sense.

16.02.2026 16:59 β€” πŸ‘ 1675    πŸ” 255    πŸ’¬ 49    πŸ“Œ 11
Prior as data, prior as belief, prior as soft constraint, prior as unconditional distribution in a generative model | Statistical Modeling, Causal Inference, and Social Science

Gelman writes...

"[a prior of normal(0, 10) is] a statement that you are only demanding that your method work well for problems in which the true theta is in that range"

What does he mean by "works well" here specifically

statmodeling.stat.columbia.edu/2025/05/21/p...

16.02.2026 17:05 β€” πŸ‘ 4    πŸ” 1    πŸ’¬ 2    πŸ“Œ 1
Preview
GitHub - DrCatHicks/learning-opportunities: A Claude Code skill for deliberate skill development during AI-assisted coding A Claude Code skill for deliberate skill development during AI-assisted coding - DrCatHicks/learning-opportunities

Key to efficient learning is realizing how we ACTUALLY learn, not just what FEELS like learning. I wrote a Claude Skill for some friends to help them think about this and they've liked it -- see Principles for some directions you could explore

github.com/DrCatHicks/l...

15.02.2026 15:54 β€” πŸ‘ 119    πŸ” 17    πŸ’¬ 6    πŸ“Œ 11

Can you learn AND offload? Yes. We do it all the time across our days. You just cannot do every single thing at once, and have to think about the structural support for different (sometimes competing) goals.

Self-regulation isn't new, but it is a vital skill for developers now.

15.02.2026 15:51 β€” πŸ‘ 86    πŸ” 13    πŸ’¬ 2    πŸ“Œ 3

In lieu of being able to clone me, you can have a longer download from me and from @analog-ashley.bsky.social as we march straight into hell (talking about whether people can learn in the AI era and how we're trying to show up for our communities on this)

www.changetechnically.fyi/2396236/epis...

16.02.2026 17:23 β€” πŸ‘ 28    πŸ” 10    πŸ’¬ 3    πŸ“Œ 2

Indeed. Though I'd go further:

At a certain level of wealth -- once you have enough money that you are insulated from any real consequences and can surround yourself with yes-men -- I think wealth and moral worth are actually anticorrelated

16.02.2026 09:05 β€” πŸ‘ 3    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

That wealth is not correlated with intelligence or moral worth is one the most important empirical discoveries ever made.

15.02.2026 21:39 β€” πŸ‘ 9    πŸ” 1    πŸ’¬ 2    πŸ“Œ 0

The study does not remotely say that science β€œleans hard left.” Nor does the fact that academics left the Nazis-and-child-porn website for a different website mean they’re producing biased evidence.

All of the major threats to science in 2026 come from the right.

14.02.2026 22:38 β€” πŸ‘ 796    πŸ” 102    πŸ’¬ 30    πŸ“Œ 5

i am entirely despondent about the state of australian journalism. the narrative that has emerged within a week is that the protesters were at fault, that they were rioting, and that the police were merely enforcing the law. if that's true, where's the footage of protesters rioting?

14.02.2026 23:30 β€” πŸ‘ 24    πŸ” 8    πŸ’¬ 4    πŸ“Œ 1
Post image

How do people decide whom to help when cooperation is costly and resources are limited?

Kinship and reciprocity are not the full storyβ€”research by @psycheddiego.bsky.social et al suggests shared fate (the other person’s outcomes influences one’s own) complements:

buff.ly/fRinBjR

14.02.2026 17:35 β€” πŸ‘ 5    πŸ” 2    πŸ’¬ 0    πŸ“Œ 0
Post image

#statstab #485 Bayesian ANCOVA and the ATE

Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting.

#ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals

solomonkurz.netlify.app/blog/2025-07...

13.02.2026 16:58 β€” πŸ‘ 15    πŸ” 4    πŸ’¬ 2    πŸ“Œ 1

I have been thinking a lot about this question. I saw David's YOLO'd paper, and have done similar things on my own. His paper's abstract is pasted below.

1. It's a remarkable achievement that CC can do this with minimal prompting.
2. The paper is unbelievably banal from a research standpoint.

14.02.2026 15:44 β€” πŸ‘ 68    πŸ” 13    πŸ’¬ 5    πŸ“Œ 11
Post image

β€œCorrelation does *not* imply causation” is not the last wordβ€”it’s only the beginning.

Cesar Chavez tracks how modern economics has been approaching causality (and the causality fallacy… and its reverse):

bit.ly/4cpiLsm

13.02.2026 19:51 β€” πŸ‘ 6    πŸ” 3    πŸ’¬ 3    πŸ“Œ 1

i came out to myself in 2013, and transitioned in 2016. the volume of hate directed at us over last 10 years has been a fucking nightmare that somehow never ends

13.02.2026 21:25 β€” πŸ‘ 12    πŸ” 4    πŸ’¬ 0    πŸ“Œ 0

Within philosophy of mind and cognitive science, ​functionalism is the doctrine that mental states are individuated by their causal role, while computationalism is the doctrine that cognition/mentation consists in symbol manipulation. These accounts have implications for what sorts of systems we can

13.02.2026 22:06 β€” πŸ‘ 30    πŸ” 7    πŸ’¬ 5    πŸ“Œ 0
Post image

Get out of that comfort zone and take a (calculated) risk!

Discomfort is good for you, argues @michelletennant:

buff.ly/j7lhdBE

14.02.2026 06:53 β€” πŸ‘ 1    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0
What to do in 2026: Your statistics diary | Statistical Modeling, Causal Inference, and Social Science

What to do in 2026: Your statistics diary
statmodeling.stat.columbia.edu/2026/02/14/w...

14.02.2026 14:05 β€” πŸ‘ 4    πŸ” 2    πŸ’¬ 0    πŸ“Œ 2
Post image 14.02.2026 15:51 β€” πŸ‘ 42    πŸ” 4    πŸ’¬ 1    πŸ“Œ 1

Just think what a salmon on a bike could do!

14.02.2026 15:51 β€” πŸ‘ 56    πŸ” 5    πŸ’¬ 2    πŸ“Œ 1
lame joke about Valentine’s Day and bash script

lame joke about Valentine’s Day and bash script

Meme from a friend:

14.02.2026 16:21 β€” πŸ‘ 14    πŸ” 2    πŸ’¬ 1    πŸ“Œ 0
Post image

Babbage, 1830, discussing the problem that scientists selectively report findings that they want to be true.

Confirmation bias is a strong human tendency. This is why we need to design science in a way that prevents conformation bias from leading us away from the truth.

14.02.2026 16:32 β€” πŸ‘ 42    πŸ” 14    πŸ’¬ 3    πŸ“Œ 1
A comic image featuring a text conversation where one person says "I love you," and the other replies, "Do you have evidence to support your statement?"

A comic image featuring a text conversation where one person says "I love you," and the other replies, "Do you have evidence to support your statement?"

Happy Valentine's Day, #philsky

14.02.2026 16:33 β€” πŸ‘ 18    πŸ” 3    πŸ’¬ 1    πŸ“Œ 0
Post image

Could life have begun with simpler molecules than we once thought? A new paper in @science.org by @edogia.bsky.social shows that a tiny RNA catalyst can self-replicate itself, suggesting that life may have been easier to emerge than expected. Getting closer. www.biorxiv.org/content/10.1...

14.02.2026 11:52 β€” πŸ‘ 45    πŸ” 18    πŸ’¬ 1    πŸ“Œ 1

😳 What?

12.02.2026 22:20 β€” πŸ‘ 5    πŸ” 2    πŸ’¬ 1    πŸ“Œ 0
Women remain underrepresented in leadership, particularly in traditionally masculine work settings. At the same time, the visibility of this imbalance has led to growing calls for diversifying leadership. This research examines how both men and women contribute to the preservation or disruption of gender inequality in masculine organizational contexts. Men remain the gatekeepers of changeβ€”deciding who rises to the top and under what conditionsβ€”while women face the strategic dilemma of fitting in by downplaying inequality (supporting the status quo, sometimes called β€˜queen bee behaviour’) or β€˜rocking the boat’ by advocating social change (challenging the status quo). Across five experimental studies (totalΒ N = 887), we examined how evaluators assessed male and female leadership candidates who either supported or challenged the status quo. Results revealed that although men favoured female over male candidates, they consistently preferred women who reinforced the status quo over those who advocated equality. By contrast, male candidates who supported the status quo were penalized, and female evaluators showed no such preferences. These findings highlight subtle mechanisms through which gendered power dynamics are maintained, underscoring both the strategic trade-offs women must navigate to advance and the conditional nature of men's support for gender equality.

Women remain underrepresented in leadership, particularly in traditionally masculine work settings. At the same time, the visibility of this imbalance has led to growing calls for diversifying leadership. This research examines how both men and women contribute to the preservation or disruption of gender inequality in masculine organizational contexts. Men remain the gatekeepers of changeβ€”deciding who rises to the top and under what conditionsβ€”while women face the strategic dilemma of fitting in by downplaying inequality (supporting the status quo, sometimes called β€˜queen bee behaviour’) or β€˜rocking the boat’ by advocating social change (challenging the status quo). Across five experimental studies (totalΒ N = 887), we examined how evaluators assessed male and female leadership candidates who either supported or challenged the status quo. Results revealed that although men favoured female over male candidates, they consistently preferred women who reinforced the status quo over those who advocated equality. By contrast, male candidates who supported the status quo were penalized, and female evaluators showed no such preferences. These findings highlight subtle mechanisms through which gendered power dynamics are maintained, underscoring both the strategic trade-offs women must navigate to advance and the conditional nature of men's support for gender equality.

Women Leaders

"Women are expected to promote gender equality and show communal behaviour, yet those who do so...are less likely to be selected for leadership."

Derks et al. (2026). Don't rock the boat! Do men prefer women leaders who support the status quo?

doi.org/10.1111/bjso...

12.02.2026 22:41 β€” πŸ‘ 28    πŸ” 8    πŸ’¬ 2    πŸ“Œ 1

@arnav7 is following 20 prominent accounts