Abhinav Kumar
@akumar03.bsky.social
<Causality | Ph.D. Candidate @mit | Physics>
I narrate (probably approximately correct) causal stories.
Past: Research Fellow @MSFTResearch
Website: abhinavkumar.info
I narrate (probably approximately correct) causal stories.
Past: Research Fellow @MSFTResearch
Website: abhinavkumar.info
Reposted by Abhinav Kumar
TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧵
🧵
The official home of the Python Programming Language
www.python.org
October 27, 2025 at 2:47 PM
TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧵
🧵
Reposted by Abhinav Kumar
A nice variant of the kernel two-sample test. arxiv.org/abs/2510.11853
Sketch of the idea: The MMD is the core of a commonly used nonparametric test for distribution testing. It works by embedding distributions into a RKHS and comparing their mean embeddings. [+]
Sketch of the idea: The MMD is the core of a commonly used nonparametric test for distribution testing. It works by embedding distributions into a RKHS and comparing their mean embeddings. [+]
A Martingale Kernel Two-Sample Test
The Maximum Mean Discrepancy (MMD) is a widely used multivariate distance metric for two-sample testing. The standard MMD test statistic has an intractable null distribution typically requiring costly...
arxiv.org
October 15, 2025 at 11:51 PM
A nice variant of the kernel two-sample test. arxiv.org/abs/2510.11853
Sketch of the idea: The MMD is the core of a commonly used nonparametric test for distribution testing. It works by embedding distributions into a RKHS and comparing their mean embeddings. [+]
Sketch of the idea: The MMD is the core of a commonly used nonparametric test for distribution testing. It works by embedding distributions into a RKHS and comparing their mean embeddings. [+]
Reposted by Abhinav Kumar
As a grad student, the biologist Yitzhi “Patrick” Cai helped program 𝘌. 𝘤𝘰𝘭𝘪 bacteria to become a biosensor for arsenic contamination in drinking water. Today, he is leading a global effort to build the first-ever synthetic eukaryotic genome. www.quantamagazine.org/hes-gleaning...
September 30, 2025 at 8:04 PM
As a grad student, the biologist Yitzhi “Patrick” Cai helped program 𝘌. 𝘤𝘰𝘭𝘪 bacteria to become a biosensor for arsenic contamination in drinking water. Today, he is leading a global effort to build the first-ever synthetic eukaryotic genome. www.quantamagazine.org/hes-gleaning...
Reposted by Abhinav Kumar
We have two new mentees who are offering their time via office hours! Please show Sandeep Silwal and Kevin Tian some love and sign up to meet them!
let-all.com/officehours....
let-all.com/officehours....
September 30, 2025 at 9:00 PM
We have two new mentees who are offering their time via office hours! Please show Sandeep Silwal and Kevin Tian some love and sign up to meet them!
let-all.com/officehours....
let-all.com/officehours....
Reposted by Abhinav Kumar
The paper this talk is based on is quite impressive arxiv.org/abs/2507.04441 one of those cases where you see direct real actionable insight using the categorical hammer.
September 25, 2025 at 2:21 PM
The paper this talk is based on is quite impressive arxiv.org/abs/2507.04441 one of those cases where you see direct real actionable insight using the categorical hammer.
Reposted by Abhinav Kumar
Today at IAS, I gave a 2 hr 15 mins lecture on why TIME[t] is in SPACE[√(t log t)]. You can watch it here!
www.youtube.com/watch?v=ThLv...
www.youtube.com/watch?v=ThLv...
Simulating Time With Square-Root Space (And With Details) - Ryan Williams
YouTube video by Institute for Advanced Study
www.youtube.com
September 23, 2025 at 8:22 PM
Today at IAS, I gave a 2 hr 15 mins lecture on why TIME[t] is in SPACE[√(t log t)]. You can watch it here!
www.youtube.com/watch?v=ThLv...
www.youtube.com/watch?v=ThLv...
Reposted by Abhinav Kumar
Aligning an AI with human preferences might be hard. But there is more than one AI out there, and users can choose which to use. Can we get the benefits of a fully aligned AI without solving the alignment problem? In a new paper we study a setting in which the answer is yes.
September 19, 2025 at 12:14 PM
Aligning an AI with human preferences might be hard. But there is more than one AI out there, and users can choose which to use. Can we get the benefits of a fully aligned AI without solving the alignment problem? In a new paper we study a setting in which the answer is yes.
Reposted by Abhinav Kumar
Sergei Drozdov has published his nice proof using hyperbolic simplexes of the necessary and sufficient condition on the radii of spheres that sit inside and outside a Euclidean simplex in any dimension.
www.sciencedirect.com/science/arti...
www.sciencedirect.com/science/arti...
Egan conjecture holds
Given a Euclidean simplex of dimension n⩾2 let its radii of inscribed and circumscribed spheres be r and R, and the distance between the centers of th…
www.sciencedirect.com
September 3, 2025 at 9:47 PM
Sergei Drozdov has published his nice proof using hyperbolic simplexes of the necessary and sufficient condition on the radii of spheres that sit inside and outside a Euclidean simplex in any dimension.
www.sciencedirect.com/science/arti...
www.sciencedirect.com/science/arti...
Reposted by Abhinav Kumar
The workshops focused on (in chronological order):
- Variational Inference (youtube.com/playlist?lis...)
- Optimal Transport (youtube.com/playlist?lis...)
- Parallel Computing (youtube.com/playlist?lis...), and
- Computational Physics (youtube.com/playlist?lis...).
- Variational Inference (youtube.com/playlist?lis...)
- Optimal Transport (youtube.com/playlist?lis...)
- Parallel Computing (youtube.com/playlist?lis...), and
- Computational Physics (youtube.com/playlist?lis...).
September 2, 2025 at 12:13 PM
The workshops focused on (in chronological order):
- Variational Inference (youtube.com/playlist?lis...)
- Optimal Transport (youtube.com/playlist?lis...)
- Parallel Computing (youtube.com/playlist?lis...), and
- Computational Physics (youtube.com/playlist?lis...).
- Variational Inference (youtube.com/playlist?lis...)
- Optimal Transport (youtube.com/playlist?lis...)
- Parallel Computing (youtube.com/playlist?lis...), and
- Computational Physics (youtube.com/playlist?lis...).
Reposted by Abhinav Kumar
The mathematician Lingrui Ge recently helped find a new way to understand the solutions of almost-periodic functions, important equations that appear in quantum physics. The work has helped cement an intriguing connection between number theory and physics. www.quantamagazine.org/ten-martini-...
August 30, 2025 at 7:45 PM
The mathematician Lingrui Ge recently helped find a new way to understand the solutions of almost-periodic functions, important equations that appear in quantum physics. The work has helped cement an intriguing connection between number theory and physics. www.quantamagazine.org/ten-martini-...
Reposted by Abhinav Kumar
Pretty cool: the "Fundamental Examples" of independence structures in Non-Commutative Probability.
August 26, 2025 at 8:11 AM
Pretty cool: the "Fundamental Examples" of independence structures in Non-Commutative Probability.
Reposted by Abhinav Kumar
Carlos Cinelli, Avi Feller, Guido Imbens, Edward Kennedy, Sara Magliacane, Jose Zubizarreta
Challenges in Statistics: A Dozen Challenges in Causality and Causal Inference
https://arxiv.org/abs/2508.17099
Challenges in Statistics: A Dozen Challenges in Causality and Causal Inference
https://arxiv.org/abs/2508.17099
August 26, 2025 at 5:56 AM
Carlos Cinelli, Avi Feller, Guido Imbens, Edward Kennedy, Sara Magliacane, Jose Zubizarreta
Challenges in Statistics: A Dozen Challenges in Causality and Causal Inference
https://arxiv.org/abs/2508.17099
Challenges in Statistics: A Dozen Challenges in Causality and Causal Inference
https://arxiv.org/abs/2508.17099
Reposted by Abhinav Kumar
Adi Shamir's advice to young researchers:
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
March 26, 2025 at 11:24 AM
Adi Shamir's advice to young researchers:
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
Reposted by Abhinav Kumar
This week's post is about why spherical cows are physics' mascot ⚛️🧪
open.substack.com/pub/nirmalya...
open.substack.com/pub/nirmalya...
An Ode to the Spherical Cow
How Imperfect Models Drive Scientific Discovery
open.substack.com
August 17, 2025 at 10:58 PM
This week's post is about why spherical cows are physics' mascot ⚛️🧪
open.substack.com/pub/nirmalya...
open.substack.com/pub/nirmalya...
Reposted by Abhinav Kumar
Big fan of this perspective:
May 7, 2025 at 6:46 PM
Big fan of this perspective:
Reposted by Abhinav Kumar
This is about one of my greatest inspirations. It would mean a lot to me if you gave it a watch
Lessons from Paula Harris / by Sophie Huiberts
YouTube video by Mixed Integer Programming
www.youtube.com
August 11, 2025 at 11:06 AM
This is about one of my greatest inspirations. It would mean a lot to me if you gave it a watch
Reposted by Abhinav Kumar
Regardless of whether you plan to use them in applications, everyone should learn about Gaussian processes, and Bayesian methods. They provide a foundation for reasoning about model construction and all sorts of deep learning behaviour that would otherwise appear mysterious.
August 9, 2025 at 2:42 PM
Regardless of whether you plan to use them in applications, everyone should learn about Gaussian processes, and Bayesian methods. They provide a foundation for reasoning about model construction and all sorts of deep learning behaviour that would otherwise appear mysterious.
Reposted by Abhinav Kumar
After a bit of a summer pause, I'm back to making episodes. In this episode, I explain the notion of confounding, and clarify why confounders should not be thought of as alternate explanations of an observed effect.
youtu.be/kAgS7cltBhM
youtu.be/kAgS7cltBhM
E5: What Confounding Really Is
YouTube video by Causal Foundations
youtu.be
August 8, 2025 at 10:18 PM
After a bit of a summer pause, I'm back to making episodes. In this episode, I explain the notion of confounding, and clarify why confounders should not be thought of as alternate explanations of an observed effect.
youtu.be/kAgS7cltBhM
youtu.be/kAgS7cltBhM
Reposted by Abhinav Kumar
Randomized trials (RCTs) help evaluate if deploying AI/ML systems actually improves outcomes (e.g., survival rates in a healthcare context).
But AI/ML systems can change: Do we need a new RCT every time we update the model? Not necessarily, as we show in our UAI paper! arxiv.org/abs/2502.09467
But AI/ML systems can change: Do we need a new RCT every time we update the model? Not necessarily, as we show in our UAI paper! arxiv.org/abs/2502.09467
July 23, 2025 at 2:10 PM
Randomized trials (RCTs) help evaluate if deploying AI/ML systems actually improves outcomes (e.g., survival rates in a healthcare context).
But AI/ML systems can change: Do we need a new RCT every time we update the model? Not necessarily, as we show in our UAI paper! arxiv.org/abs/2502.09467
But AI/ML systems can change: Do we need a new RCT every time we update the model? Not necessarily, as we show in our UAI paper! arxiv.org/abs/2502.09467
Reposted by Abhinav Kumar
I had a great time presenting "It's Time to Say Goodbye to Hard Constraints" at the Flatiron Institute. In this talk, I describe a philosophy for model construction in machine learning. Video now online! www.youtube.com/watch?v=LxuN...
It's Time to Say Goodbye to Hard (equivariance) Constraints - Andrew Gordon Wilson
YouTube video by LoG Meetup NYC
www.youtube.com
July 22, 2025 at 7:28 PM
I had a great time presenting "It's Time to Say Goodbye to Hard Constraints" at the Flatiron Institute. In this talk, I describe a philosophy for model construction in machine learning. Video now online! www.youtube.com/watch?v=LxuN...
Reposted by Abhinav Kumar
Armin Keki\'c, Jan Schneider, Dieter B\"uchler, Bernhard Sch\"olkopf, Michel Besserve
Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
https://arxiv.org/abs/2507.14901
Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
https://arxiv.org/abs/2507.14901
July 22, 2025 at 4:45 AM
Armin Keki\'c, Jan Schneider, Dieter B\"uchler, Bernhard Sch\"olkopf, Michel Besserve
Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
https://arxiv.org/abs/2507.14901
Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
https://arxiv.org/abs/2507.14901
Reposted by Abhinav Kumar
Hirahara, Illango, and Loff posted on the arXiv a lovely result, showing that determining the communication complexity of a function f is NP-hard. A fundamental question first asked by Yao in '79. The proof is very clean and elegant. A fun read for the weekend!
arxiv.org/pdf/2507.104...
arxiv.org/pdf/2507.104...
arxiv.org
July 19, 2025 at 11:28 AM
Hirahara, Illango, and Loff posted on the arXiv a lovely result, showing that determining the communication complexity of a function f is NP-hard. A fundamental question first asked by Yao in '79. The proof is very clean and elegant. A fun read for the weekend!
arxiv.org/pdf/2507.104...
arxiv.org/pdf/2507.104...
Reposted by Abhinav Kumar
📢ICML alert: In the afternoon poster session, I'll present our paper: "Contextures: Representations from Contexts".
Our central argument: "Representations are learned from the association between input 𝑋 and context variable 𝐴"
📍East: E-1708, July 15, 4.30-7pm
📜 openreview.net/pdf?id=4GZwFPz…
Our central argument: "Representations are learned from the association between input 𝑋 and context variable 𝐴"
📍East: E-1708, July 15, 4.30-7pm
📜 openreview.net/pdf?id=4GZwFPz…
July 15, 2025 at 5:54 PM
📢ICML alert: In the afternoon poster session, I'll present our paper: "Contextures: Representations from Contexts".
Our central argument: "Representations are learned from the association between input 𝑋 and context variable 𝐴"
📍East: E-1708, July 15, 4.30-7pm
📜 openreview.net/pdf?id=4GZwFPz…
Our central argument: "Representations are learned from the association between input 𝑋 and context variable 𝐴"
📍East: E-1708, July 15, 4.30-7pm
📜 openreview.net/pdf?id=4GZwFPz…
Reposted by Abhinav Kumar
Here's 7 habits to start this week! youtu.be/cqjf4DJyAaA?...
7 Simple Daily Habits That Will Change Your Life (Stoic-Inspired)
YouTube video by Daily Stoic
youtu.be
July 14, 2025 at 2:51 PM
Here's 7 habits to start this week! youtu.be/cqjf4DJyAaA?...
Reposted by Abhinav Kumar
Learning Actionable Counterfactual Explanations in Large State Spaces
Keziah Naggita, Matthew Walter, Avrim Blum
Action editor: Taylor Killian
https://openreview.net/forum?id=tXnVRpRlR8
#actions #explanations #features
Keziah Naggita, Matthew Walter, Avrim Blum
Action editor: Taylor Killian
https://openreview.net/forum?id=tXnVRpRlR8
#actions #explanations #features
July 15, 2025 at 12:08 AM
Learning Actionable Counterfactual Explanations in Large State Spaces
Keziah Naggita, Matthew Walter, Avrim Blum
Action editor: Taylor Killian
https://openreview.net/forum?id=tXnVRpRlR8
#actions #explanations #features
Keziah Naggita, Matthew Walter, Avrim Blum
Action editor: Taylor Killian
https://openreview.net/forum?id=tXnVRpRlR8
#actions #explanations #features