Pingsheng Li
@pingshengli.bsky.social
CS PhD, Neuro Undergrad @mcgillu, @Mila_Quebec, LiNC Lab @tyrell_turing, Prev Neuro-X @EPFL, intern @GatsbyUCL, Neuro AI Convergence DL&RL.
Reposted by Pingsheng Li
New preprint! 🧠🤖
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
June 6, 2025 at 5:40 PM
New preprint! 🧠🤖
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
Reposted by Pingsheng Li
Here's our latest preprint on neural decoders for spiking data. Stay tuned for the code (and hopefully, some exciting follow-ups)!
New preprint! 🧠🤖
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
June 6, 2025 at 5:54 PM
Here's our latest preprint on neural decoders for spiking data. Stay tuned for the code (and hopefully, some exciting follow-ups)!
Reposted by Pingsheng Li
I’m looking for interns to join our lab for a project on foundation models in neuroscience.
Funded by @ivado.bsky.social and in collaboration with the IVADO regroupement 1 (AI and Neuroscience: ivado.ca/en/regroupem...).
Interested? See the details in the comments. (1/3)
🧠🤖
Funded by @ivado.bsky.social and in collaboration with the IVADO regroupement 1 (AI and Neuroscience: ivado.ca/en/regroupem...).
Interested? See the details in the comments. (1/3)
🧠🤖
AI and Neuroscience | IVADO
ivado.ca
November 7, 2025 at 1:52 PM
I’m looking for interns to join our lab for a project on foundation models in neuroscience.
Funded by @ivado.bsky.social and in collaboration with the IVADO regroupement 1 (AI and Neuroscience: ivado.ca/en/regroupem...).
Interested? See the details in the comments. (1/3)
🧠🤖
Funded by @ivado.bsky.social and in collaboration with the IVADO regroupement 1 (AI and Neuroscience: ivado.ca/en/regroupem...).
Interested? See the details in the comments. (1/3)
🧠🤖
There might be many kinds of intelligence, but there is only one reality.
How could they not, ultimately, converge?
How could they not, ultimately, converge?
May 1, 2025 at 6:42 AM
There might be many kinds of intelligence, but there is only one reality.
How could they not, ultimately, converge?
How could they not, ultimately, converge?
Reposted by Pingsheng Li
Come by Poster 068 to learn about why comp neuro studies should use exponentiated gradient descent!
March 29, 2025 at 5:22 PM
Come by Poster 068 to learn about why comp neuro studies should use exponentiated gradient descent!
Reposted by Pingsheng Li
I turned 45 this weekend, which means I can finally become the man I was always destined to be:
youtu.be/KF2X1o0FGA0?...
youtu.be/KF2X1o0FGA0?...
He's Hip. He's Cool. He's 45!
YouTube video by Rosemary's Baby
youtu.be
January 20, 2025 at 9:17 PM
I turned 45 this weekend, which means I can finally become the man I was always destined to be:
youtu.be/KF2X1o0FGA0?...
youtu.be/KF2X1o0FGA0?...
100 gradient updates per year 🥹
December 15, 2024 at 9:11 PM
100 gradient updates per year 🥹
A must read for neuroAI!!!!!!!!!!
After CNN, other great feature of the brain that specific has computational advantage.
After CNN, other great feature of the brain that specific has computational advantage.
Why does #compneuro need new learning methods? ANN models are usually trained with Gradient Descent (GD), which violates biological realities like Dale’s law and log-normal weights. Here we describe a superior learning algorithm for comp neuro: Exponentiated Gradients (EG)! 1/12 #neuroscience 🧪
October 31, 2024 at 6:45 PM
A must read for neuroAI!!!!!!!!!!
After CNN, other great feature of the brain that specific has computational advantage.
After CNN, other great feature of the brain that specific has computational advantage.