Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore how cryptographic primitives can be leveraged to achieve significant improvements in algorithmic time complexity in this advanced computer science seminar. Learn about the innovative concept of "Trapdoor Matrix Distributions" and discover how standard cryptographic assumptions can be used to design algorithms that are asymptotically faster than existing solutions while maintaining correctness. Examine the first uniform reduction from worst-case to approximate and average-case matrix multiplication with optimal parameters, and understand breakthrough applications including worst-case to average-case reductions for matrix inversion and other linear operations. Delve into fast general-purpose dimension reductions and explore how these techniques can accelerate inference time in classification models. Gain insights into cutting-edge research that bridges cryptography and computational complexity theory, demonstrating how cryptographic tools can serve non-cryptographic objectives beyond their traditional roles in eliminating randomness and reducing interaction.
