Vin Bhaskara
@vinbhaskara.github.io
Sr. Applied Scientist, Borealis AI. Prev’ly, Samsung AI Toronto. Vector Inst Scholar. MSc Univ of Toronto, B.Tech. IIT-Guwahati. Opinions my own.
🔗 vinbhaskara.github.io
📍 Montréal, Canada 🇨🇦
🔗 vinbhaskara.github.io
📍 Montréal, Canada 🇨🇦
Pinned
Vin Bhaskara
@vinbhaskara.github.io
· Mar 2
Skye 🐈 agrees with Prof. Hinton about the looming AI Doomsday and is visibly upset :/
November 24, 2025 at 12:38 AM
Skye 🐈 agrees with Prof. Hinton about the looming AI Doomsday and is visibly upset :/
Hands down the best introduction to all the Differential Geometry and Riemannian Geometry you will ever need to understand General Relativity - this is gold: www.youtube.com/watch?v=7G4S...
Lecture 1: Topology (International Winter School on Gravity and Light 2015)
YouTube video by The WE-Heraeus International Winter School on Gravity and Light
www.youtube.com
November 23, 2025 at 6:01 AM
Hands down the best introduction to all the Differential Geometry and Riemannian Geometry you will ever need to understand General Relativity - this is gold: www.youtube.com/watch?v=7G4S...
So many recent vision papers using VQ-VAE... Time to give it another read: arxiv.org/abs/1711.00937
Neural Discrete Representation Learning
Learning useful representations without supervision remains a key challenge in machine learning. In this paper, we propose a simple yet powerful generative model that learns such discrete representati...
arxiv.org
November 4, 2025 at 3:42 PM
So many recent vision papers using VQ-VAE... Time to give it another read: arxiv.org/abs/1711.00937
Finally! Got the indices right in deriving the Geodesic Equation with the Christoffel symbols expressed in the standard form! Phew (and wohoo!)!!! Tl;dr. Finding the shortest path under a general metric is an insanely complicated thing computationally, quite unintuitive but true!
November 1, 2025 at 8:46 PM
Finally! Got the indices right in deriving the Geodesic Equation with the Christoffel symbols expressed in the standard form! Phew (and wohoo!)!!! Tl;dr. Finding the shortest path under a general metric is an insanely complicated thing computationally, quite unintuitive but true!
N-th attempt at deriving the Geodesic equation without messing up the tensor indices... so close to having the right Christoffel tensor expression now but it's still missing terms:( 😮💨
October 30, 2025 at 1:34 PM
N-th attempt at deriving the Geodesic equation without messing up the tensor indices... so close to having the right Christoffel tensor expression now but it's still missing terms:( 😮💨
My PhD students are now out on the academic job market. Please give them jobs (and tuna). Ty.
October 28, 2025 at 12:59 PM
My PhD students are now out on the academic job market. Please give them jobs (and tuna). Ty.
Reposted by Vin Bhaskara
“Terrible things are happening outside. Poor helpless people are being dragged out of their homes. Families are torn apart. Men, women, and children are separated. Children come home from school to find that their parents have disappeared.”
Diary of Anne Frank
January 13, 1943
Diary of Anne Frank
January 13, 1943
October 4, 2025 at 8:17 PM
“Terrible things are happening outside. Poor helpless people are being dragged out of their homes. Families are torn apart. Men, women, and children are separated. Children come home from school to find that their parents have disappeared.”
Diary of Anne Frank
January 13, 1943
Diary of Anne Frank
January 13, 1943
I've a new law based on my observations of my cats. The center of mass of all cats in the household is a random walk around the pet owner's coordinates.
September 23, 2025 at 9:19 PM
I've a new law based on my observations of my cats. The center of mass of all cats in the household is a random walk around the pet owner's coordinates.
A metric g(x) describes a flat intrinsic geometry iff there exists a coordinate transformation T(x) such that the metric under the transformation becomes an Identity matrix for all x.
i.e., transpose(T(x)) * g(x) * T(x) = Identity matrix; for all x.
i.e., transpose(T(x)) * g(x) * T(x) = Identity matrix; for all x.
September 22, 2025 at 5:25 PM
A metric g(x) describes a flat intrinsic geometry iff there exists a coordinate transformation T(x) such that the metric under the transformation becomes an Identity matrix for all x.
i.e., transpose(T(x)) * g(x) * T(x) = Identity matrix; for all x.
i.e., transpose(T(x)) * g(x) * T(x) = Identity matrix; for all x.
Yes. And, Pineapple on Pizza ❤️
September 11, 2025 at 9:05 PM
Yes. And, Pineapple on Pizza ❤️
Wrap Text >>>>> No Wrap. Always! Any global setting to make this the default @vscode.dev ?!
September 11, 2025 at 9:04 PM
Wrap Text >>>>> No Wrap. Always! Any global setting to make this the default @vscode.dev ?!
July 11, 2025 at 1:45 AM
discipline >>>>>>>> inspiration or motivation
July 2, 2025 at 6:07 PM
discipline >>>>>>>> inspiration or motivation
consistency >>>> bursts of productivity
July 2, 2025 at 6:05 PM
consistency >>>> bursts of productivity
a palet breton a day keeps the therapist away...
May 20, 2025 at 2:16 PM
a palet breton a day keeps the therapist away...
Reposted by Vin Bhaskara
Distrobox is so much cleaner and lightweight, I love it! Finally time to move away from the Docker memory hog
May 8, 2025 at 8:53 PM
Distrobox is so much cleaner and lightweight, I love it! Finally time to move away from the Docker memory hog
A place of worship to reflect on what matters. Frederick Banting lives on through his gift of Insulin to the world that he sold to UofT for $1. Some live once, and some others live forever through their actions... #BantingAndBest #ThisIsCANADA #Never51
May 4, 2025 at 1:01 PM
A place of worship to reflect on what matters. Frederick Banting lives on through his gift of Insulin to the world that he sold to UofT for $1. Some live once, and some others live forever through their actions... #BantingAndBest #ThisIsCANADA #Never51
ChatGPT grounded in Real-TIme US Clinical Trials Data:
Want to find out which clinical trials are ongoing for a drug you take?
Curious about the sample size, inclusion criteria, or outcomes for a condition affecting someone you know?
👇 (🧵 1/3)
Want to find out which clinical trials are ongoing for a drug you take?
Curious about the sample size, inclusion criteria, or outcomes for a condition affecting someone you know?
👇 (🧵 1/3)
April 19, 2025 at 4:29 AM
ChatGPT grounded in Real-TIme US Clinical Trials Data:
Want to find out which clinical trials are ongoing for a drug you take?
Curious about the sample size, inclusion criteria, or outcomes for a condition affecting someone you know?
👇 (🧵 1/3)
Want to find out which clinical trials are ongoing for a drug you take?
Curious about the sample size, inclusion criteria, or outcomes for a condition affecting someone you know?
👇 (🧵 1/3)
Reposted by Vin Bhaskara
I don't always post on Twitter. But when I do...
April 9, 2025 at 8:39 PM
I don't always post on Twitter. But when I do...
Xubuntu >> Lubuntu factoring in both performance and basic usability on a low-resource machine. My desperate attempt to bring a old Chromebook 4 to life... finally worked!
April 1, 2025 at 12:27 AM
Xubuntu >> Lubuntu factoring in both performance and basic usability on a low-resource machine. My desperate attempt to bring a old Chromebook 4 to life... finally worked!
"if it's not calibrated it's not probability but just a number between 0 and 1"
March 22, 2025 at 4:00 PM
"if it's not calibrated it's not probability but just a number between 0 and 1"
The modern ‘aha!’ moment for Masked Language Models (MLMs) trained on large-scale data is few-shot learning—the ability to generalize well with very few task-specific labeled examples. This is analogous to how generalist conversational AI became the defining breakthrough for LLMs.
March 22, 2025 at 3:54 PM
The modern ‘aha!’ moment for Masked Language Models (MLMs) trained on large-scale data is few-shot learning—the ability to generalize well with very few task-specific labeled examples. This is analogous to how generalist conversational AI became the defining breakthrough for LLMs.
Reposted by Vin Bhaskara
Tonight an unbelievable **34,000 people** gathered for our Denver rally to take on billionaires and win our country back.
This was the largest political gathering in Denver since Obama in 2008.
Also bigger than the 2024 DNC.
And the largest ever rally in Bernie’s career (and obviously, mine too).
This was the largest political gathering in Denver since Obama in 2008.
Also bigger than the 2024 DNC.
And the largest ever rally in Bernie’s career (and obviously, mine too).
March 22, 2025 at 1:29 AM
Tonight an unbelievable **34,000 people** gathered for our Denver rally to take on billionaires and win our country back.
This was the largest political gathering in Denver since Obama in 2008.
Also bigger than the 2024 DNC.
And the largest ever rally in Bernie’s career (and obviously, mine too).
This was the largest political gathering in Denver since Obama in 2008.
Also bigger than the 2024 DNC.
And the largest ever rally in Bernie’s career (and obviously, mine too).
# python wizardry:
> a = [[]] * 5
> print(a)
[[], [], [], [], []]
> a[0].append(1)
> print(a)
[[1], [1], [1], [1], [1]]
# instead use list comprehension for expected result:
> a = [[] for i in range(5)]
> a[0].append(1)
> print(a)
[[1], [], [], [], []]
> a = [[]] * 5
> print(a)
[[], [], [], [], []]
> a[0].append(1)
> print(a)
[[1], [1], [1], [1], [1]]
# instead use list comprehension for expected result:
> a = [[] for i in range(5)]
> a[0].append(1)
> print(a)
[[1], [], [], [], []]
March 20, 2025 at 5:03 AM
# python wizardry:
> a = [[]] * 5
> print(a)
[[], [], [], [], []]
> a[0].append(1)
> print(a)
[[1], [1], [1], [1], [1]]
# instead use list comprehension for expected result:
> a = [[] for i in range(5)]
> a[0].append(1)
> print(a)
[[1], [], [], [], []]
> a = [[]] * 5
> print(a)
[[], [], [], [], []]
> a[0].append(1)
> print(a)
[[1], [1], [1], [1], [1]]
# instead use list comprehension for expected result:
> a = [[] for i in range(5)]
> a[0].append(1)
> print(a)
[[1], [], [], [], []]