And yesterday I took a break from the meeting to head to Caltech to meet my PhD advisor Ravin Bhatt was was visiting

Very gratifying to present today at the APS meeting in Anaheim California

From blog.samaltman.com/three-observ..., if I understand correctly, this implies that if you increase classical resources by 10x, you only get an incremental bump in classical AI performance.

I'm guessing they have no experimentalists on their 'test and evaluation team'

Thank you President Biden.

Even if a quantum model trained today does not meet the baseline performance to beat a classical deep learning model, its parameters are still useful in jumpstarting the training of a future larger model! As an example, see the training below on MNIST:

Picture the entanglement net expanding with each new node being optimized as it is added.

I introduce 'bit-bit encoding' in which the most predictive bits from a dataset are mapped to specific qubits, and the notion of an `entanglement net' with nodes representing parameterized entangling operations. The entanglement net can grow as the number of available qubits and gates increases.