Tong Liang

AI Practitioner in Healthcare

[Tong Liang @universeesee]

I'm excited to share that our paper, "Statistically inferred neuronal connections in subsampled neural networks strongly correlate with spike train covariances", was just published in Physical Review E!

https://t.co/Bsedz6s4YY

[Zidan Yang | 杨紫丹 @Zidan_Yang]

Excited to share my first paper from Inagaki Lab @MPFNeuro!

How does the “timer” in our brain control movement? We discovered specialized computational roles of the frontal cortex and striatum that implement such an action timer⏱️

https://t.co/qSUp6TRTbA

[Tong Liang @universeesee]

I'm grateful to my colleagues at @SBUNeurobiology for their support of this work. Please feel free to reach out if you have any questions or would like to discuss this research in the context of causal inference in neuroscience! :)
#Neuroscience #inference #causalinference

[Tong Liang @universeesee]

I'm excited to share that our paper, "Statistically inferred neuronal connections in subsampled neural networks strongly correlate with spike train covariances", was just published in Physical Review E!

https://t.co/Bsedz6s4YY

[Tong Liang @universeesee]

Caution should be exercised when interpreting statistically inferred connections as causal relationships, particularly when working with subsampled neural data.

[Tong Liang @universeesee]

Excited to see our new preprint out recently! @Braden_Brinkman and I studied how the commonly used maximum likelihood estimation plays a role in neuronal connection inference. See the threads below for details!

[Tong Liang @universeesee]

It's been an incredible 4.5 years working with Dr. Braden Brinkman. Without any exaggeration, Braden is the most supportive advisor I would ever expect. Thank you, @Braden_Brinkman, for being an extraordinary advisor! ❤️ Looking forward to the new chapters ahead of me!

[Tong Liang @universeesee]

@xiong_yanyu Thanks Yanyu!!

[Tong Liang @universeesee]

Ever wondered how animals 'ought' to navigate their environments & produce offspring? We used a mechanistic evolutionary dynamics approach to study the evolution of innate behavioral strategies in populations of simulated agents. Summary :thread: (1/8) https://t.co/wKfXjFVBh0

[Tong Liang @universeesee]

In the future, we hope to study agents with more complex sensory processing and interactions with the environment, and test whether the behavioral strategies predicted by pre-defining fitness functions tend to agree with the outcomes of population dynamics models. (8/8)

[Tong Liang @universeesee]

We were not able to derive an effective fitness function in this regime, suggesting perhaps the necessity of considering multi-agent objective functions when interactions between agents are common. (7/8)