Scientific computing is increasingly incorporating the advancements in
machine learning and the ability to work with large amounts of data. At the
same time, machine learning models are becoming increasingly sophisticated and
exhibit many features often seen in scientific computing, stressing the
capabilities of machine learning frameworks. Just as the disciplines of
scientific computing and machine learning have shared common underlying
infrastructure in the form of numerical linear algebra, we now have the
opportunity to further share new computational infrastructure, and thus ideas,
in the form of Differentiable Programming. We describe Zygote, a Differentiable
Programming system that is able to take gradients of general program
structures. We implement this system in the Julia programming language. Our
system supports almost all language constructs (control flow, recursion,
mutation, etc.) and compiles high-performance code without requiring any user
intervention or refactoring to stage computations. This enables an...

more |
pdf
| html
hackernewsbot:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/7Qts4wrsdo

newsycombinator:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/KaxnXxhykm

HNTweets:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing: https://t.co/zao7Ap1HxC Comments: https://t.co/lJVoWcGWmt

octonion:
A Differentiable Programming System to Bridge ML and Scientific Computing - https://t.co/OPZZIWwz6J

michael_nielsen:
I hope this is part of a trend exploring more and more in this direction. As @karpathy said, "gradient descent is a better programmer than you": https://t.co/2t1Sp0YOcz https://t.co/FhxtH9YUe2

newsycombinator:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/KaxnXxhykm

newsyc250:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/ylVaDjURgn (https://t.co/OLEttM3vd1)

BrundageBot:
Zygote: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing. Mike Innes, Alan Edelman, Keno Fischer, Chris Rackauckus, Elliot Saba, Viral B Shah, and Will Tebbutt https://t.co/l1fve59SfR

lobsters:
A Differentiable Programming System to Bridge Machine Learning and Scientific Computing
via @as1ndu
https://t.co/LBzkVlVerL #ai #compsci
https://t.co/zuyuD5NM14

ireneista:
Holy shit. The machines are coming for our job.
(Okay, not *quite*. But on first read, it seems like a huge step. We didn't previously know that the field of differentiable programming existed.)
https://t.co/niS0Tcm6sg

newsyc200:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/YGIEXeJDbt (https://t.co/wMBSByZ7kW)

jreuben1:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/m38DjLbpJk

ChrisRackauckas:
#julialang is enabling scientific machine learning. Take at a the paper submitted to NeurIPS, which showcases quantum machine learning and neural stochastic differential equation (neural SDE) as applications. Many talks at #juliacon will explore this idea
https://t.co/HjzCqXOEOu

hn_frontpage:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing
L: https://t.co/bykofKZqX0
C: https://t.co/9FDYOgQGW9

watrcoolr:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/p7HItrbw47

brendanjmeade:
I want to believe (in @JuliaLanguage): https://t.co/NgcTqaxXdR

M157q_News_RSS:
A Differentiable Programming System to Bridge ML and Scientific Computing
https://t.co/Dk30roQxJG
Article URL: https://t.co/Dk30roQxJG Comments URL: https://t.co/Xsr95w3WGO Points: 113 # Comments: 12

M157q_News_RSS:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing
https://t.co/Dk30roQxJG
Article URL: https://t.co/Dk30roQxJG Comments URL: https://t.co/Xsr95w3WGO Points: 100 # Comments: 10

RebelDevNews:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/70ebxpWGeD

hackernewsj:
∂P：MLと科学計算を橋渡しする微分可能プログラミングシステム https://t.co/TwhgCIXItt

SimonDanisch:
If I hooked anyone with my @curry_on_conf talk (https://t.co/P0uJrLwve6) here is a great further read: https://t.co/cHdjksPQ3S It explains all the interesting bits I couldn't dive into! (How ML, Scientific Computing and AD all come together in #JuliaLang )

hackernewsfeed:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/tYeelWKbj2

arxivml:
"Zygote: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing",
Mike Innes, Alan…
https://t.co/we9bBx8HCO

stephenrra:
Zygote: A system for ∂P capabilities in Julia: https://t.co/qi7jPGQxeO

hackernews100:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/9dfijLKTxf

jargnar:
This was such a wonderful read! "A differentiable programming system to bridge machine learning and scientific computing" https://t.co/1YgR0Ud9Nr

osibot:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/2zJDwMzVZc

tmcpro:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/T17ksJ7Y2t

emsal1863:
https://t.co/7ist4tTvtD
"∂P: A Differentiable Programming System to Bridge ML and Scientific Computing"
This is some exciting stuff. The lstm example in sec 3.1 is crazy

hacker_news_hir:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing : https://t.co/HdFbUL8XT5 Comments: https://t.co/okBaDf6dQe

HighSNHN:
A Differentiable Programming System to Bridge ML and Scientific Computing: https://t.co/bvNgxqHFCy ( https://t.co/GgCdjirtlw )

hackernewsrobot:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/EObtu6cTs2

hereticreader:
Zygote: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing - https://t.co/9D1COPFWPY https://t.co/Z7bgDp7jTr

betterhn100:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/o13787ZX5i

TheNisargDave:
Worth reading: A differentiable programming system for modern machine learning and scientific computing applications,
https://t.co/O5LwH1IQVg

rogerluorl18:
Zygote paper is out with a VQE demo made with Yao! I'm looking forward to the talk. Nice work @MikeJInnes
https://t.co/ruSQkk2CJN

_mangesh_tekale:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/dCyoKw0Ao9

_mangesh_tekale:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/dCyoKw0Ao9

arxiv_cs_LG:
Zygote: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing. Mike Innes, Alan Edelman, Keno Fischer, Chris Rackauckus, Elliot Saba, Viral B Shah, and Will Tebbutt https://t.co/SyfNUr58eT

mohapsat:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/gj3VM18ts9 #AI #news #tech

e3pi:
#ml #DeepLearning #julialang #c01e
"..Zygote -a differentiable programming system that is able to take gradients of general program..in the Julia
https://t.co/qmBKBY0zts https://t.co/qWgJiukH1O

deeplearning4j:
A Differentiable Programming System to Bridge Machine Learning and Scientific Computing
https://t.co/qElMO1Ks8M

betterhn50:
54 – ∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/kHiOmeQ7iO

gfrphy:
so Zygote paper has been published. I'm too excited with this ^^
https://t.co/ROWdcQ5yjJ
Zygote is one of the most ambitious Julia Project after JuMP project.

Memoirs:
Zygote: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing. https://t.co/GyE0DKQC1w

Hn150:
A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/3CuWqDbraF (https://t.co/YTOf3XPRGk)

exaday:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/Pn0h4Nfycb

wasnhierlos:
∂P: A Differentiable Programming System to Bridge ML and Scientific Computing - https://t.co/lJ0npQpmbI

aficionado:
A Differentiable Programming System to Bridge Machine Learning and Scientific Computing https://t.co/Ys9C8VqvA0

SantchiWeb:
RT @aficionado: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing https://t.co/Ys9C8VqvA0

maymounkov:
RT @newsycombinator: ∂P: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/KaxnXxhykm

hailaioverlords:
RT @mohapsat: A Differentiable Programming System to Bridge ML and Scientific Computing https://t.co/gj3VM18ts9 #AI #news #tech

en_zxteloiv:
RT @deeplearning4j: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing
https://t.co/qElMO1Ks8M

Stargazers: 19

Subscribers: 5

Subscribers: 5

Forks: 6

Open Issues: 1

Open Issues: 1

None.

Sample Sizes : None.

Authors: 7

Total Words: 5870

Unqiue Words: 2362

Functional programming languages, such as Haskell, enable simple, concise,
and correct-by-construction hardware development. HTCC compiles a subset of
Haskell to Handel-C language with hardware output. Moreover, HTCC generates
VHDL, Verilog, EDIF, and SystemC programs. The design of HTCC compiler includes
lexical, syntax and semantic analyzers. HTCC automates a transformational
derivation methodology to rapidly produce hardware that maps onto Field
Programmable Gate Arrays (FPGAs) . HTCC is generated using ANTLR
compiler-compiler tool and supports an effective integrated development
environment. This paper presents the design rationale and the implementation of
HTCC. Several sample generations of first-class and higher-order functions are
presented. In-addition, a compilation case-study is presented for the XTEA
cipher. The investigation comprises a thorough evaluation and performance
analysis. The targeted FPGAs include Cyclone II, Stratix IV, and Virtex-6 from
Altera and Xilinx.

more |
pdf
| html
arxiv_org:
HTCC: Haskell to Handel-C Compiler. https://t.co/R4fHSZiOc4 https://t.co/4l9ueqrHPW

None.

None.

Sample Sizes : None.

Authors: 2

Total Words: 4696

Unqiue Words: 1849

In this paper, we take a pervasively effectful (in the style of ML) typed
lambda calculus, and show how to extend it to permit capturing pure expressions
with types. Our key observation is that, just as the pure simply-typed lambda
calculus can be extended to support effects with a monadic type discipline, an
impure typed lambda calculus can be extended to support purity with a comonadic
type discipline. We establish the correctness of our type system via a simple
denotational model, which we call the capability space model. Our model
formalizes the intuition common to systems programmers that the ability to
perform effects should be controlled via access to a permission or capability,
and that a program is capability-safe if it performs no effects that it does
not have a runtime capability for. We then identify the axiomatic categorical
structure that the capability space model validates, and use these axioms to
give a categorical semantics for our comonadic type system. We then give an
equational theory (substitution and the...

more |
pdf
| html
None.

jjcarett2:
Beautiful paper of the day: Recovering Purity with Comonads and Capabilities, by Vikraman Choudhury and Neel Krishnaswami. https://t.co/wR6f4c7t2R. A delightful read.

arxiv_cslo:
Recovering Purity with Comonads and Capabilities https://t.co/Ng3Z8qKrVM

None.

None.

Sample Sizes : None.

Authors: 2

Total Words: 0

Unqiue Words: 0

Foundational verification allows programmers to build software which has been
empirically shown to have high levels of assurance in a variety of important
domains. However, the cost of producing foundationally verified software
remains prohibitively high for most projects,as it requires significant manual
effort by highly trained experts. In this paper we present Proverbot9001 a
proof search system using machine learning techniques to produce proofs of
software correctness in interactive theorem provers. We demonstrate
Proverbot9001 on the proof obligations from a large practical proof project,the
CompCert verified C compiler,and show that it can effectively automate what was
previously manual proofs,automatically solving 15.77% of proofs in our test
dataset. This corresponds to an over 3X improvement over the prior state of the
art machine learning technique for generating proofs in Coq.

more |
pdf
| html
None.

None.

None.

Sample Sizes : None.

Authors: 4

Total Words: 12139

Unqiue Words: 3139

The actor model is popular for many types of server applications. Efficient
snapshotting of applications is crucial in the deployment of pre-initialized
applications or moving running applications to different machines, e.g for
debugging purposes. A key issue is that snapshotting blocks all other
operations. In modern latency-sensitive applications, stopping the application
to persist its state needs to be avoided, because users may not tolerate the
increased request latency. In order to minimize the impact of snapshotting on
request latency, our approach persists the application's state asynchronously
by capturing partial heaps, completing snapshots step by step. Additionally,
our solution is transparent and supports arbitrary object graphs. We prototyped
our snapshotting approach on top of the Truffle/Graal platform and evaluated it
with the Savina benchmarks and the Acme Air microservice application. When
performing a snapshot every thousand Acme Air requests, the number of slow
requests ( 0.007% of all requests) with latency...

more |
pdf
| html
None.

None.

None.

Sample Sizes : None.

Authors: 4

Total Words: 11220

Unqiue Words: 2817

Assert is a website where the best academic papers on arXiv (computer science, math, physics), bioRxiv (biology), BITSS (reproducibility), EarthArXiv (earth science), engrXiv (engineering), LawArXiv (law), PsyArXiv (psychology), SocArXiv (social science), and SportRxiv (sport research) bubble to the top each day.

Papers are scored (in real-time) based on how verifiable they are (as determined by their Github repos) and how interesting they are (based on Twitter).

To see top papers, follow us on twitter @assertpub_ (arXiv), @assert_pub (bioRxiv), and @assertpub_dev (everything else).

To see beautiful figures extracted from papers, follow us on Instagram.

*Tracking 160,434 papers.*

Sort results based on if they are interesting or reproducible.

Interesting

Reproducible