11,545 Hits in 3.9 sec

Probabilistic Alias Analysis for Parallel Programming in SSA Forms [article]

Mohamed A. El-Zawawy, Mohammad N. Alanazi
2014 arXiv   pre-print
This paper presents a new probabilistic approach for alias analysis of parallel programs.  ...  Probabilistic versions of these results, in which result elements are associated with occurrence probabilities, are required in optimizations techniques of modern compilers.  ...  The authors acknowledge the support (grants numbers 340918 & 330911) of the deanship of scientific research of Al Imam Mohammad Ibn Saud Islamic University (IMSIU).  ... 
arXiv:1405.4401v1 fatcat:6ihi5q3yqrasjm5zvyj2gd7clu

From high-level inference algorithms to efficient code [article]

Rajan Walia, Praveen Narayanan, Jacques Carette, Sam Tobin-Hochstadt, Chung-chieh Shan
2019 arXiv   pre-print
We show how probabilistic programs that directly and concisely express these desired inference algorithms can be compiled while maintaining efficiency.  ...  We introduce new transformations that turn high-level probabilistic programs with arrays into pure loop code.  ...  our compilation pipeline is the efficient execution of array inference algorithms expressed as probabilistic programs denoting conditional distributions.  ... 
arXiv:1805.06562v5 fatcat:kta3djopvnhs5fwpv7bro2zjqy

Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming [article]

Guillaume Baudart, Javier Burroni, Martin Hirzel, Louis Mandel, Avraham Shinnar
2021 arXiv   pre-print
Stan is a probabilistic programming language that is popular in the statistics community, with a high-level syntax for expressing probabilistic models.  ...  Stan differs by nature from generative probabilistic programming languages like Church, Anglican, or Pyro.  ...  ,e n ),e) in k Comprehensive compilation of statements Theorem 3. 3 . 3 For all Stan programs p, the semantics of the source program is equal to the semantics of the compiled program up to a constant  ... 
arXiv:1810.00873v5 fatcat:3lcvh6vr6rbxhaszuuuvjpsjuu

Augur: a Modeling Language for Data-Parallel Probabilistic Inference [article]

Jean-Baptiste Tristan, Daniel Huang, Joseph Tassarotti, Adam Pocock, Stephen J. Green, Guy L. Steele Jr
2014 arXiv   pre-print
In this paper, we present a probabilistic programming language and compiler for Bayesian networks designed to make effective use of data-parallel architectures such as GPUs.  ...  Probabilistic programming addresses this problem by allowing a user to specify the model and having a compiler automatically generate an inference procedure for it.  ...  The key feature of probabilistic programming is separation of concerns: the user specifies what needs to be learned by describing a probabilistic model, while the compiler automatically generates the how  ... 
arXiv:1312.3613v2 fatcat:vi6klyjuhrbelh5hr3yqau6vsy

Finally, a Polymorphic Linear Algebra Language (Pearl)

Amir Shaikhha, Lionel Parreaux, Michael Wagner
2019 European Conference on Object-Oriented Programming  
This design enables us to change the behaviour of arithmetic operations to express matrix algebra, graph algorithms, logical probabilistic programs, and differentiable programs.  ...  Crucially, the polymorphic design of Pilatus allows us to use multistage programming and rewrite-based optimisation to recover the performance of specialised code, supporting fixed sized matrices, algebraic  ...  1 We used Scala as the implementation language for Pilatus, but other programming languages with support for lambda expressions and multi-stage programming could be used as well; most of the techniques  ... 
doi:10.4230/lipics.ecoop.2019.25 dblp:conf/ecoop/ShaikhhaP19 fatcat:uw35z2shozcs3lghnngzenfrwe

Probabilistic source-level optimisation of embedded programs

Björn Franke, Michael O'Boyle, John Thomson, Grigori Fursin
2005 SIGPLAN notices  
Efficient implementation of DSP applications is critical for many embedded systems.  ...  Optimising C compilers for embedded processors largely focus on code generation and instruction scheduling which, with their growing maturity, are providing diminishing returns.  ...  Deriving efficient program transformation sequences, however, is a complex task.  ... 
doi:10.1145/1070891.1065922 fatcat:2uihscs3xve77lclwhmjo5ojcq

Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro [article]

Du Phan, Neeraj Pradhan, Martin Jankowiak
2019 arXiv   pre-print
NumPyro is a lightweight library that provides an alternate NumPy backend to the Pyro probabilistic programming language with the same modeling interface, language primitives and effect handling abstractions  ...  In particular, NumPyro provides an iterative formulation of the No-U-Turn Sampler (NUTS) that can be end-to-end JIT compiled, yielding an implementation that is much faster than existing alternatives in  ...  Memory Efficiency To tackle the issue of memory efficiency, we will use an array S to store only even numbered nodes z k at index i = BitCount(k).  ... 
arXiv:1912.11554v1 fatcat:ulnibhbmsratfcbc7ffstre6le

A technological review of the FORTRAN I compiler

F. E. Allen
1982 Proceedings of the June 7-10, 1982, national computer conference on - AFIPS '82  
The FORTRAN I compiler functions and organizations are described and shown to form the basis for many of the techniques used in modern compilers.  ...  The FORTRAN program in Figure 1 , moved array B to array A in a double nest of DO loops.  ...  (Section 4) To perform the control flow analysis and identify (probabilistically) the relative frequency of program regions. 3.  ... 
doi:10.1145/1500774.1500875 dblp:conf/afips/Allen82 fatcat:xz4zadcanfdyvc46yglaznnkfe

Probabilistic Data Programming with ENFrame

Dan Olteanu, Sebastiaan J. van Schaik
2014 IEEE Data Engineering Bulletin  
The thesis of this work is that one can build powerful and useful probabilistic data programming frameworks that leverage existing work on probabilistic databases.  ...  This paper overviews ENFrame, a programming framework for probabilistic data.  ...  language. • ENFrame exploits the structure of queries and programs for efficient inference; it relies on SPROUT for query processing on probabilistic data [16, 9] .  ... 
dblp:journals/debu/OlteanuS14 fatcat:ctwdvnshv5ey7ceiwflmamqkuy

Data Layout Optimization for the LCC Compiler

Viktor Shamparov, Murad Neiman-Zade
2021 Proceedings of the Institute for System Programming of RAS  
The approach consists of three parts. The first part is generalizing two methods of estimating cache miss amount and choosing more applicable one in the compiler.  ...  The second part is finding an applicable solution for the problem of cache miss amount minimization.  ...  Thus, to make compiled program use cache memory efficiently, the compiler must improve these two programs' properties.  ... 
doi:10.15514/ispras-2021-33(3)-4 fatcat:fswwc673efhl3m2o3fzfeuxrui

Symbolic Exact Inference for Discrete Probabilistic Programs [article]

Steven Holtzen, Todd Millstein, Guy Van den Broeck
2019 arXiv   pre-print
To do this, we first compile probabilistic programs to a symbolic representation.  ...  The computational burden of probabilistic inference remains a hurdle for applying probabilistic programming languages to practical problems of interest.  ...  Science Foundation, IIS-1657613, IIS-1633857, CCF-1837129, DARPA XAI grant N66001-17-2-4032, NEC Research, and a gi from Intel. e authors would like to thank Joe Qian for assistance with the development of  ... 
arXiv:1904.02079v3 fatcat:3n2dajixgfggzlzy5fh6ifqqzu

Compilation for fast calculation over pedigrees

P. Szolovits
1992 Cytogenetic and Genome Research  
Efficient computation of probabilistic relationships over family pedigrees is an important tool for a variety of problems in genetics, including genetic counseling and linkage analysis.  ...  A number of typical compiler optimization techniques can be applied to the resulting program to speed up computations, typically at the expense of (much) more space and more compiletime analysis.  ...  Efficient computation of probabilistic relationships over family pedigrees is an important tool for a variety of problems in genetics, including genetic counseling and linkage analysis.  ... 
doi:10.1159/000133226 pmid:1737481 fatcat:stun2jgn6bgdzkvyuz2vhplpqq

Compiling probabilistic, bio-inspired circuits on a field programmable analog array

Bo Marr, Jennifer Hasler
2014 Frontiers in Neuroscience  
A field programmable analog array (FPAA) is presented as an energy and computational efficiency engine: a mixed mode processor for which functions can be compiled at significantly less energy costs using  ...  The Gillespie algorithm is simulated to show the utility of this system by calculating the trajectory of a biological system computed stochastically with this probabilistic hardware where over a 127X performance  ...  The authors would also like to thank the organizers and participants of the 2008 Telluride Neuromorphic Workshop and the Institute for Neuromorphic Engineering. Dr.  ... 
doi:10.3389/fnins.2014.00086 pmid:24847199 pmcid:PMC4019887 fatcat:qm4nmotk5fekdomywd2lvjjdbu


Dan Olteanu, Sebastiaan J. Van Schaik
2016 ACM Transactions on Database Systems  
The program is then interpreted probabilistically by ENFrame. The realisation of ENFrame required novel contributions along several directions.  ...  of program variables.  ...  PDFs of program variables can be further processed, e.g. by a probabilistic DBMS.  ... 
doi:10.1145/2877205 fatcat:uak2a3wvqbdaroacav2eyvlts4

ENFrame: A Platform for Processing Probabilistic Data [article]

Sebastiaan J. van Schaik and Dan Olteanu and Robert Fink
2013 arXiv   pre-print
The program is then interpreted probabilistically by ENFrame. The realisation of ENFrame required novel contributions along several directions.  ...  of program variables.  ...  The semantics of ENFrame programs is based on a unified probabilistic interpretation of the entire processing pipeline from the input data to the program result.  ... 
arXiv:1309.0373v1 fatcat:mqv5xq53srhgrf4k6hiw2qxnxy
« Previous Showing results 1 — 15 out of 11,545 results