Speeding up sparse convolutions: placement experience

[fjebaker]

Grant will tell us about what he did during his placement.

[grant-m-s]

For those interested the talk will cover topics including:

• How Pytorch performs Convolutions under-the-hood
• Implementation of novel methods that outperform default Pytorch implementations
• Code developed using CUDA and C++
• How LiDAR is processed in autonomous vehicles
• Optimising processes for computing on the edge (high compute, low power)
• What is the process of getting a patent for a project (from the engineering perspective)

[fjebaker]

Thanks @grant-m-s for the presentation. Is there a chance the slides can be made (either publicly or privately) available?

[grant-m-s]

Not sure the extent at which the slides can be shared so i'd rather stay on the safe side and not share them out, sorry about that

[fjebaker]

No problem!