-
We would like to use the autodiff library with custom arithmetic types, such as posits, valids, unums, Elias, fixed-point, etc. so that autodiff can be used in environments where these types are instr…
-
**Github username:** --
**Twitter username:** --
**Submission hash (on-chain):** 0x2e5384ed5a8dd0b973cc6bd44aed557f48e829e32e756078df273572bdec2664
**Severity:** high
**Description:**
**Description*…
-
The Lattigo library (https://github.com/tuneinsight/lattigo) is an open source library for lattice-based homomorphic encryption in Go. At v5, they have organized the library into many packages, from l…
asraa updated
3 months ago
-
@jschall once proposed to add those to the set of native DSDL types:
> Can we use a floating point standard that matches IEEE 754 for 16, 32, 64 and 128-bit floats, but also provides for arbitrary …
-
Hi, thanks for your excellent work to improve the scalability and performance of HLS, which is very helpful for hardware accelerator development.
After looking through your paper and code, I am won…
-
Could you please expose `genConstructor`, `genMatConstructor`, `genQuatConstructor` to library user? For example I would like to write a library for fixed point arithmetic and use my new datatypes wit…
-
Could you elaborate a little bit more what are the main optimizations of that project that make it faster on 486 and other non pentium CPUs ? README does not tell much.
AFAIK a lot of CPU time in Q…
-
# Fixed point truncation
## Conceptual overview
Truncate fixed point register to lower precision.
### Current status
Needs to be done manually and is very errorprone.
## Proposal
###…
-
The functions of FixedPointCoordinate happily cast to floats and use the FPU for arithmetic operations.
Goal is to
- replace _everything_ with fixed point computation,
- be faster, and
- be as prec…
-
Suyog Gupta, Ankur Agrawal, Kailash Gopalakrishnan, Pritish Narayanan
Training of large-scale deep neural networks is often constrained by the available computational resources. We study the effect…