anirban sarkar

Postdoc@Cold Spring Harbor Lab, previously postdoc@MIT, PhD@IIT Hyderabad

[Peter Koo @pkoo562]

Excited to share an update to D3 (DNA Discrete Diffusion) — an application of score-entropy discrete diffusion model for regulatory genomics!

🧬 Paper: https://t.co/7H8DDqES2B

(See thread below 👇) (1/n)

[Peter Koo @pkoo562]

Excited to share new work on “Designing DNA With Tunable Regulatory Activity Using Discrete Diffusion.” We extend Score Entropy Discrete Diffusion to regulatory DNA! Work led by @ani_itsme.

paper: https://t.co/qK47aTrJMC

[anirban sarkar @ani_itsme]

@anindya_ece @EaseMyTrip @British_Airways Why don't you take care of customer's money and satisfaction if things don't go as planned? Please show some kindness as well as professionalism here and arrange for the refund ASAP.

[anirban sarkar @ani_itsme]

@ExpediaHelp I mailed twice for the refund status but didn't get any satisfactory reply. Please do the needful ASAP.

[anirban sarkar @ani_itsme]

@Expedia @ExpediaIN @ExpediaAsia @ExpediaHelp @JALairlines @JAL_Official_jp
It's been more than 4 months I didn't get the refund, even the booking was cancelled successfully and got confirmation of refund within 4-6 weeks. Not sure which side it is stuck with.

[anirban sarkar @ani_itsme]

Is taking more time normal? When can I actually expect the refund?

[anirban sarkar @ani_itsme]

@PajamaStew Thank you for putting the whole incident down so nicely. I could relate to this more as I experienced a death of an elderly figure in my family recently.

[Grant Sanderson @3blue1brown]

What is a convolution? What is a "Gaussian Kernel"?

(This was the main animation used at the intro of last week's convolutions lecture).

[Frank Nielsen @FrnkNlsn]

Common formula for statistical (dis)similarities between any two densities of a same exponential family (incl. Gaussian, Beta, Dirichlet): Implement those formula *easily* from legacy statistical library APIs.
Slides: https://t.co/cTdZIV1zaJ
Report: https://t.co/H0475IkB8t

[Judea Pearl @yudapearl]

I am in receipt of a comprehensive paper on the role of causality in vision https://t.co/X1rO0QdS8a Although I find the section on "causal Bayes Nets" a bit incoherent, there is a lot I can learn from it, especially the interplay with naive physics and object-oriented reasoning.

[Judea Pearl @yudapearl]

Very pleased to find a powerful completeness result in the theory of transportability: https://t.co/NByCvF2NiY
This time the heterogeneous data sets can have missing components and the targets can be group-specific causal effect.

[Dmitry Kobak @hippopedoid]

Did you know that the optimal ridge penalty λ in linear regression can be *negative*? It's always strictly positive when n>p. Or when cov(x)=I. Or when true β is random. But here we argue that it can be zero or even negative when p>>n: https://t.co/LzS6L5xrbP. HOW?! [1/n]

[Frank Nielsen @FrnkNlsn]

New method to calculate and express the Kullback-Leibler divergence between densities of an exponential family (normals, gamma, etc.) by expressing the KL as an averaged sum of log density ratio. Bypass integral calculations and use of Bregman generators https://t.co/Blx3eQlmlG

[Judea Pearl @yudapearl]

Speaking of DL and causal inference (CI) and keeping with our commitment to on-line education, I am retweeting here a video-ed lecture on the subject https://t.co/JnSDTceXEY. It's more than a year old, but covers many of the questions raised here, squarely and transparently.

[Judea Pearl @yudapearl]

For the many students of economics who express frustration with the way a problem known as “bad control” is evaded, if not mishandled in econometrics, our "Crash Course" in now accessible as Technical Report: https://t.co/eQnCbLDsuE
Enjoy and teach your professor.
#Bookofwhy

[Judea Pearl @yudapearl]

The reason I said "insurmountable obstacle" is that it cannot be solved by model-free ML methods. The examples in https://t.co/dEPwcuLbPS show that the same shift in probability may require two different repairs, depending on which structure caused the shift.#Bookofwhy

[Dan Roy @roydanroy]

In "Mondrian Kernel", we showed the Mondrian process can approximate the Laplace kernel. Using the Mondrian process <> STIT connection, O'Reilly and Tran have now uncovered whole classes of kernels that can be approximate by stochastic geometries. Beauty! https://t.co/xVDHpZycte

[Judea Pearl @yudapearl]

I cannot overemphasize the importance of page 24 for
understanding: "intervention", "parents", "invariance"
and other notions, often treated informally. Chapter 3 later explicates the ramifications of this one page, and gives us backdoor, do-calculus and much more, #Bookofwhy

[anirban sarkar @ani_itsme]

@sachin_rt who is ur favourite in todays game???

[anirban sarkar @ani_itsme]

@ImZaheer no need to worry...we d supporters r always with u all...we blv dat india will win d nxt two matches nd we jst cnt wait to see dat