synapse
play

SYNAPSE with Metasketches POPL 2016 James Bornholt Emina Torlak - PowerPoint PPT Presentation

Mar 21, 2024 •579 likes •1.4k views

f a i c t t r A C o m p * * l e t t n e e * A t * s i W s E n e L o l l C C P D * o O * e c s u P u m e E e R n * o t e v t d y s a E * a d l e u t a Optimizing Synthesis

  1. f a i c t t r A C o m p * * l e t t n e e * A t * s i W s E n e L o l l C C P D * o O * e c s u P u m e E e R n * o t e v t d y s a E * a d l e u t a Optimizing Synthesis SYNAPSE with Metasketches POPL 2016 James Bornholt Emina Torlak Dan Grossman Luis Ceze University of Washington

  2. f a i c t t r A C o m p * * l e t t n e e * A t * s i W s E n e L o l l C C P D * o O * e c s u P u m e E e R n * o t e v t d y s a E * a d l e u t a Optimizing Synthesis SYNAPSE with Metasketches d e t a m o POPL 2016 t u A r o f e t a m i x o r p p A James Bornholt g n i m m a r Emina Torlak g o r P Dan Grossman Luis Ceze University of Washington

  3. Approximate computing in the future

  4. Approximate computing in the future Precise Program

  5. Approximate computing in the future Precise Acceptability Program Criteria

  6. Approximate computing in the future Approximation Specification

  7. Approximate computing in the future Approximation Specification Magical Approximate Compiler

  8. Approximate computing in the future Approximation Specification Magical Approximate Compiler Approximate Program

  9. Synthesis finds programs automatically Specification Program Synthesis Program

  10. Synthesis finds programs automatically f ( x ) = 4 x + 1 Specification Program Synthesis Program

  11. Synthesis finds programs automatically f ( x ) = 4 x + 1 Specification Program Synthesis Program 4*x + 1

  12. Synthesis finds programs automatically f ( x ) = 4 x + 1 Specification Program Synthesis Program 4*x + 1 (x << 2) + 1

  13. What if synthesis doesn’t work? Specification Program Synthesis Program

  14. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Specification Program Synthesis Program

  15. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Specification Program Synthesis Program

  16. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Restricts the space of Syntactic candidate programs Specification Template (e.g. sketch, CFG) Program Synthesis Program

  17. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Restricts the space of Syntactic candidate programs Specification Template (e.g. sketch, CFG) Syntax-Guided Synthesis Program

  18. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Restricts the space of Syntactic candidate programs Specification Template (e.g. sketch, CFG) Syntax-Guided Synthesis Linear search Program Binary search

  19. What if synthesis doesn’t work? f ( L, k ) = Search ( L, k ) Restricts the space of Syntactic candidate programs Specification Template (e.g. sketch, CFG) Syntax-Guided Synthesis Linear search Program Binary search 7.1 MB program

  20. What if synthesis doesn’t work? Selects the solution f ( L, k ) = Search ( L, k ) with minimal cost Cost Syntactic Specification Function Template Syntax-Guided Synthesis Linear search Program Binary search 7.1 MB program

  21. What if synthesis doesn’t work? Selects the solution f ( L, k ) = Search ( L, k ) with minimal cost Cost Syntactic Specification Function Template Optimal Synthesis Linear search Program Binary search

  22. What if synthesis doesn’t work? Selects the solution f ( L, k ) = Search ( L, k ) with minimal cost Cost Syntactic Specification Function Template Optimal Synthesis Program

  23. Metasketches o ff er control over synthesis Specification Optimal Synthesis Program

  24. Metasketches o ff er control over synthesis Guides the search towards good solutions Specification Metasketch Optimal Synthesis Program

  25. Metasketches o ff er control over synthesis Guides the search towards good solutions 1. Metasketches Specification Metasketch 2. Solving Optimal Synthesis 3. Results Program

  26. Metasketches o ff er control over synthesis Guides the search towards good solutions 1. Metasketches Specification Metasketch 2. Solving Optimal Synthesis 3. Results Program

  27. Metasketches Explicit control over search strategy

  28. Synthesizing straight-line SSA programs max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) o2 = ?? op (?? a , ?? a , ?? a ) return o2

  29. Synthesizing straight-line SSA programs Operations: +, -, <, if, … max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) o2 = ?? op (?? a , ?? a , ?? a ) return o2

  30. Synthesizing straight-line SSA programs Operations: +, -, <, if, … max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) o2 = ?? op (?? a , ?? a , ?? a ) return o2 Inputs + earlier outputs

  31. Synthesizing straight-line SSA programs max(x, y): o1 = >(x, y) o2 = ?? op (?? a , ?? a , ?? a ) return o2

  32. Synthesizing straight-line SSA programs max(x, y): o1 = >(x, y) o2 = ite(o1, x, y) return o2

  33. Synthesizing straight-line SSA programs max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) o2 = ?? op (?? a , ?? a , ?? a ) return o2

  34. Synthesizing straight-line SSA programs max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 o2 = ?? op (?? a , ?? a , ?? a ) return o2

  35. Synthesizing straight-line SSA programs max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  36. Synthesizing straight-line SSA programs max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  37. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  38. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  39. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 SAT, cost 5 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  40. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 SAT, cost 5 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 SAT, cost 4 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 S 4 S 5 S 6 S 7 S 8 S 9 …

  41. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 2 SAT, cost 5 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 SAT, cost 4 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 4 4 4 3 3 3 S 4 S 5 S 6 S 7 S 8 S 9 …

  42. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 Termination of optimal max(x, y): synthesis o1 = ?? op (?? a , ?? a , ?? a ) S 2 SAT, cost 5 o2 = ?? op (?? a , ?? a , ?? a ) return o2 max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 SAT, cost 4 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 4 4 4 3 3 3 S 4 S 5 S 6 S 7 S 8 S 9 …

  43. Optimal synthesis could be unbounded max(x, y): Unbounded space of o1 = ?? op (?? a , ?? a , ?? a ) candidate programs UNSAT S 1 return o1 Termination of optimal max(x, y): synthesis o1 = ?? op (?? a , ?? a , ?? a ) S 2 SAT, cost 5 o2 = ?? op (?? a , ?? a , ?? a ) return o2 Search order max(x, y): o1 = ?? op (?? a , ?? a , ?? a ) S 3 SAT, cost 4 o2 = ?? op (?? a , ?? a , ?? a ) o3 = ?? op (?? a , ?? a , ?? a ) return o3 4 4 4 3 3 3 S 4 S 5 S 6 S 7 S 8 S 9 …

  44. Metasketches control search strategy Unbounded space of candidate programs Termination of optimal synthesis Search order

  45. Metasketches control search strategy A metasketch consists of: Unbounded space of candidate programs Termination of optimal synthesis Search order

  46. Metasketches control search strategy A metasketch consists of: Unbounded space of • An ordered countable set of sketches S candidate programs Termination of optimal synthesis Search order

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Module 1: Synapses, plasticity and circuits The synapse: transfer of information 1 ms The

Module 1: Synapses, plasticity and circuits The synapse: transfer of information 1 ms The

Module 1: Synapses, plasticity and circuits The synapse: transfer of information 1 ms The synapse: transfer of information The synapse The miniature postsynaptic response (or mini) Fatt and Katz, 1952 - Remain in the presence of TTX -

859 views • 53 slides
A chemical synapse, by EM http://web.as.uky.edu/Biology/faculty/cooper/CooperRes2.htm An

A chemical synapse, by EM http://web.as.uky.edu/Biology/faculty/cooper/CooperRes2.htm An

A chemical synapse, by EM http://web.as.uky.edu/Biology/faculty/cooper/CooperRes2.htm An electrical synapse, by EM http://electroneubio.secyt.gov.ar/Bennett-neurons'electrical_synchronization.htm Purkinje neuron filled with fluorescent dye and

299 views • 12 slides
Transform Your Business IT as s a Servi vice ce Wel elcome come to Synapse apse 360 Agenda

Transform Your Business IT as s a Servi vice ce Wel elcome come to Synapse apse 360 Agenda

Transform Your Business IT as s a Servi vice ce Wel elcome come to Synapse apse 360 Agenda 11 July 2019 10:45 Welcome &amp; Introduction Steve Douglas, Sales Director, Synapse 360 10:55 Hybrid Cloud Services Overview Resilient &amp;

469 views • 16 slides
Communication Between Neurons Synapse: A specialized site of contact, and transmission of

Communication Between Neurons Synapse: A specialized site of contact, and transmission of

Communication Between Neurons Synapse: A specialized site of contact, and transmission of information between a neuron and an effector cell Anterior Motor Neuron Figure 45-5 Communication Between Neurons Chemical synapse

1.13k views • 66 slides
Rhode Island Carbon Pricing Study Cadmus Group &amp; Synapse Energy Economics, Inc. Tuesday, May

Rhode Island Carbon Pricing Study Cadmus Group &amp; Synapse Energy Economics, Inc. Tuesday, May

Policy Options Meeting Rhode Island Carbon Pricing Study Cadmus Group &amp; Synapse Energy Economics, Inc. Tuesday, May 19 th 2020 Cadmus &amp; Synapse Presenters Farrah Andersen Jesse Way Asa Hopkins Associate Senior Analyst Vice

645 views • 42 slides
Synapse Formation in the Extension of axons/axon guidance Peripheral and Central Target

Synapse Formation in the Extension of axons/axon guidance Peripheral and Central Target

Synapse formation completes the wiring of the nervous system Birth and differentiation of neurons Synapse Formation in the Extension of axons/axon guidance Peripheral and Central Target recognition Nervous System Synaptic

199 views • 7 slides
Geo-Spatial Design p g Creativity is the synapse between two normally disassociated matrices

Geo-Spatial Design p g Creativity is the synapse between two normally disassociated matrices

Specialist Meeting on Spatial Concepts in GIS and Design Santa Barbara, California Geo-Spatial Design p g Creativity is the synapse between two normally disassociated matrices of thought. g Arthur Koestler William R. Miller

940 views • 27 slides
Synapse A Microservices Architecture for Heterogeneous-Database Web Application Nicolas Viennot

Synapse A Microservices Architecture for Heterogeneous-Database Web Application Nicolas Viennot

F o r k m e o n G i t h u b Synapse A Microservices Architecture for Heterogeneous-Database Web Application Nicolas Viennot Mathias Lcuyer Jonathan Bell Roxana Geambasu Jason Nieh Columbia University Thank you for the intro.

763 views • 53 slides
Parallel vs Antiparallel Synapse Parallel vs Antiparallel Synapse Site-specific Recombination

Parallel vs Antiparallel Synapse Parallel vs Antiparallel Synapse Site-specific Recombination

Parallel vs Antiparallel Synapse Parallel vs Antiparallel Synapse Site-specific Recombination Site specific Recombination Resolvase Recombination Resolvase Recombination Cre Recombination Cre Recombination Active Site Configuration in Cre

134 views • 10 slides
Synapse Medical Services www.synapsemedical.com.au BILLING | CODING | TRANSCRIPTION |

Synapse Medical Services www.synapsemedical.com.au BILLING | CODING | TRANSCRIPTION |

Global Medical Administration Synapse Medical Services www.synapsemedical.com.au BILLING | CODING | TRANSCRIPTION | EDUCATION Expectation Reality Chennai Ops The Solution Pricing Some of the team No more coding backlogs!

376 views • 10 slides
Simulation of the Ephaptic Effect in the Cone- Horizontal Cell Synapse of the Retina Carl

Simulation of the Ephaptic Effect in the Cone- Horizontal Cell Synapse of the Retina Carl

Simulation of the Ephaptic Effect in the Cone- Horizontal Cell Synapse of the Retina Carl Gardner, Jeremiah Jones, Steve Baer, &amp; Sharon Crook School of Mathematical &amp; Statistical Sciences Arizona State University

742 views • 31 slides
An Introduction to Energy Subsidies May 16, 2019 Synapse Energy Economics, Inc. Webinar Series

An Introduction to Energy Subsidies May 16, 2019 Synapse Energy Economics, Inc. Webinar Series

An Introduction to Energy Subsidies May 16, 2019 Synapse Energy Economics, Inc. Webinar Series Doug Koplow Earth Track, Inc. 485 Massachusetts Ave. #2 Cambridge, MA 02139 (617) 661-4700 DKoplow@earthtrack.net Overview Selling the Dream:

454 views • 26 slides
How the Brain Works 1 Background: The Synapse 2 Release Signal 3 Receptor binding 4

How the Brain Works 1 Background: The Synapse 2 Release Signal 3 Receptor binding 4

How the Brain Works 1 Background: The Synapse 2 Release Signal 3 Receptor binding 4 Receptors: Excitatory: Sends signals (action potentials) Inhibitory: Blocks signals Drugs, neurotransmitters, and other ligands: Agonists: Stimulate

758 views • 27 slides
A model of pattern completion based on the CA3 recurrent synapse Jose Guzman, Michael Frotscher

A model of pattern completion based on the CA3 recurrent synapse Jose Guzman, Michael Frotscher

A model of pattern completion based on the CA3 recurrent synapse Jose Guzman, Michael Frotscher and Peter Jonas The neural structure for pattern completion Histologie du Systeme Nerveus de lHomme et des Vertebres, Vols. 1 &amp; 2. A.

339 views • 32 slides
4/5/16 April 4, 2016 Neural Networks for Motor Pattern Production Central Pattern Generator Lab

4/5/16 April 4, 2016 Neural Networks for Motor Pattern Production Central Pattern Generator Lab

4/5/16 April 4, 2016 Neural Networks for Motor Pattern Production Central Pattern Generator Lab Due Today: Sim 8 Thursday: Lab Results- Synapse 2 Plasticity; Plant Due Friday: Synapse 1 Rhythmic motor patterns important for locomotion,

437 views • 5 slides
NEST Modeling Language: A modeling language for spiking neuron and synapse models for NEST

NEST Modeling Language: A modeling language for spiking neuron and synapse models for NEST

NEST Modeling Language: A modeling language for spiking neuron and synapse models for NEST I.Blundell, D.Plotnikov, J.M.Eppler Software Engineering RWTH Aachen FZ Jlich Blundell, Plotnikov Lehrstuhl fr Software Engineering Wait, yet

251 views • 22 slides
Synaptic Learning Rules Computational Models of Neural Systems Lecture 4.1 David S. Touretzky

Synaptic Learning Rules Computational Models of Neural Systems Lecture 4.1 David S. Touretzky

Synaptic Learning Rules Computational Models of Neural Systems Lecture 4.1 David S. Touretzky October, 2019 Why Study Synaptic Plasticity? Synaptic learning rules determine the information processing capabilities of neurons. Synaptic

888 views • 40 slides
Optimizing Synthesis f a i c t t r A C o m p * * l e t t n e e * A t * s

Optimizing Synthesis f a i c t t r A C o m p * * l e t t n e e * A t * s

Optimizing Synthesis f a i c t t r A C o m p * * l e t t n e e * A t * s i W s E n e L o l l C C P D * o O * e c s u P u m e E e R n with Metasketches * o t e v t d y s a E * a d

1.88k views • 108 slides
Testing Overview Introduction (Proving and Testing) Types of Testing Unit,

Testing Overview Introduction (Proving and Testing) Types of Testing Unit,

CPSC 310 Software Engineering Testing Overview Introduction (Proving and Testing) Types of Testing Unit, Integration, Regression, System, Acceptance Testing Tactics Functional (Black Box), Structural (White Box)

982 views • 61 slides
Recent RPS Studies: Maine and Maryland Hosted by Warren Leon, Executive Director, CESA April

Recent RPS Studies: Maine and Maryland Hosted by Warren Leon, Executive Director, CESA April

CESA Webinar Recent RPS Studies: Maine and Maryland Hosted by Warren Leon, Executive Director, CESA April 15, 2020 Housekeeping Join audio: Choose Mic &amp; Speakers to use VoIP Choose Telephone and dial using the information

460 views • 31 slides
Xcel Energy Performance Metrics Stakeholder Engagement Process MN PUC Docket 17-401 AGENDA

Xcel Energy Performance Metrics Stakeholder Engagement Process MN PUC Docket 17-401 AGENDA

Xcel Energy Performance Metrics Stakeholder Engagement Process MN PUC Docket 17-401 AGENDA Welcome Overview on Utility Performance Metrics Overview of PUC Order and Stakeholder Engagement Process Q&amp;A Welcome Rolf

578 views • 30 slides
100% Containers Powered Carpooling Maxime Fouilleul Database Reliability Engineer BlaBlaCar -

100% Containers Powered Carpooling Maxime Fouilleul Database Reliability Engineer BlaBlaCar -

100% Containers Powered Carpooling Maxime Fouilleul Database Reliability Engineer BlaBlaCar - Facts &amp; Figures Infrastructure Ecosystem - 100% containers powered carpooling Todays agenda Stateful Services into containers - MariaDB as

463 views • 45 slides
The technical and mathematical challenges What are the goals that set the challenges? We want

The technical and mathematical challenges What are the goals that set the challenges? We want

The technical and mathematical challenges What are the goals that set the challenges? We want maps that can yield an atomic model with a known accuracy, and we want to visualize conformational variability. We want tomograms of cells in which we

813 views • 37 slides
Excel Modelling Approach Nick Ogden ACA Dr Josef Baker 06/09/2018 Nick Ogden: Background to the

Excel Modelling Approach Nick Ogden ACA Dr Josef Baker 06/09/2018 Nick Ogden: Background to the

Excel Modelling Approach Nick Ogden ACA Dr Josef Baker 06/09/2018 Nick Ogden: Background to the Problem The first problem I wanted to solve was that of building and maintaining the integrity of Forecasts and Budgets Building the forecast

439 views • 28 slides

More recommend