M
Machine learning (large-scale)
algorithm and implementation
280–287
with atomics and bijection
284–285
base algorithm
279–280
butterfly sum reduction
285
character recognition application
289–290
core technology
278–280
energy calculation
282
future work
290
gradient calculation
287
host-device interface
282
overview
277
performance enhancements
287–290
probability normalization
282–286
theoretical background
278–279
training overview
281
value summing
286–287
without atomics
283–284
MacMolPlt, MO visualization
8
Macro-gates
activation rate
359–361
event-driven simulation
352
future work
362–363
oblivious simulation
352–354
segmentation process
356–358
simulation phase, steps
351–355
sizing heuristics
355
MAE
Magic filters, BigDFT code
136–143
Magnetic resonance imaging (MRI) reconstruction
B-spline registration
751–752
with field inhomogeneity compensation
core method
710–713
future work
720–721
high-speed implementations
715–718
kernel functions
715–717
MATLAB toolbox
718–719
multi-GPU implementation
718
non-Cartesian sampling trajectory
710–711
overview
709–710
with parallel programming
712–713
performance analysis
719–720
SpMV
717
susceptibility-induced
711–712
SPIRiT, compressed sensing
algorithms, implementations, evaluations
727–733
approach
724–725
basic problem
723–726
Calibration Consistency operator
727–729
core methods
726–727
data consistency projection
729
overview
735
parallel imaging
726
performance analysis
733–734
soft thresholding
729
wavelet transform
729–733
Main memory access
fast spectral synthesis
95
medical image processing
746
MO computations
13
molecular electrostatics
53
VSM
465
Map step, SMO
297–298
Markov-Random-Fields (MRF), Graph Cuts
439
Massively parallel hybrid CPU-GPU clusters, BigDFT code
benefits and limitations
144–145
BLAS routines
140–143
code structure
138
convolutions
138–139
core method
135–138
CPU code
145
Daubechies wavelets
134
definition
133–134
efficiency developments
149
hybrid code performance
145–147
implementation
140
kinetic convolution and preconditioner
139
kinetic operator
136
local potential
136
magic filters
136
molecule simulation domain example
135
multiple core calculations
148
operations
136–137
overview
134
parallel distribution
147
3D operators
140–143
Massive parallel computing, genome-matching
algorithms, implementations, evaluations
176–183
basic problem
173–174
benefits and limitations
183
core methods
174–176
CUDA-CPU execution
181
CUDA kernel parameter settings
180–181
data layout
179–180
future work
183
gaps and mismatches
181–182
genome encoding
177–178
hashing
182
performance scaling
182–183
smaller threads
181
target headers
178–179
target processing
180
Math Kernel Library (MKL)
MATLAB toolbox
MaxEnt model
base algorithm
280
character recognition application
289–290
performance enhancements
287–290
training overview
281
VSM
465–466
Matrix Market Coordinate Formal (MMCF), MRI reconstruction
717
Matrix multiplication
atlas construction
781
facial animation
417
FFT
629
for large-scale facial deformation
418
machine learning
287–290
Mersenne Twister MT19937
237
RNA folding
205–206
Maximum entropy (MaxEnt), machine learning
algorithm and implementation
280–287
with atomics and bijection
284–285
base algorithm
279–280
butterfly sum reduction
285
core technology
278–280
energy calculation
282
future work
290
gradient calculation
287
host-device interface
282
overview
277
performance enhancements
287–290
probability normalization
282–286
theoretical background
278–279
training overview
281
value summing
286–287
without atomics
283–284
maxScore, SW algorithm
156
Max-throughput problem, gate-level task scheduling
372
MDTS
Mean absolute error (MAE), CT reconstruction parameters
698
MEAs
Medical imaging
GPU computing status
644
Memory-access patterns
brain connectivity reconstruction
802
Chaos Game algorithm
267–269
CT image reconstruction
669–670
image de-mosaicing
585
integral image representation
523
iterated function systems
267–270
LDPC
622–623
molecular electrostatics
51–52
real-time stereo
475
RNA folding
209
temporal data mining
219
Memory access strategy, facial animation
423–424
Memory fence, Barnes Hut
n
-body algorithm
82
Memory latency
Barnes-Hut
n
-body algorithm
86
chemical informatics
23
image de-mosaicing
593
object-detection with CUDA
532–533
radiographic image simulation
823–824
temporal data mining
225
Memory tuning, chemical informatics
28–30
Mersenne Twister random number generator
and Chaos game algorithm
270
formulation
237–238
parallelization
238
performance evaluation
242–245
point generation
238–240
random walks in path tracing
406
serial implementation
239
skip-ahead algorithm
240–242
state updates
238–240
Mesh building, as spectral synthesis step
94
Mesh rendering, as spectral synthesis step
94
MEX functions, MaxEnt model
281
Minimum average correlation energy (MACE), feature-based pipeline
501–504
Minimum-phone-error (MPE) criterion, speech models
613
MKL
MNIST dataset, SVM evaluation
307–308
Molecular dynamics (MD) methods
basic problem
59–61
building blocks and implementation
65–68
calculations
69–72
density matrix
69
direct
Fock matrix
69
hardware considerations
69
matrix operations
66–68
overview
59
technology basics
61–65
Molecular orbital (MO) computations
carbon-60 multi-GPU performance
18
carbon-60 single-GPU performance
17
constant cache
13
core method
6
CPMD
59
hardware global memory cache
15
isosurfaces example
6
JIT kernel generation
17
kernel comparison
8
mathematical background
8–10
multilevel parallel decomposition
11–13
overview
5–6
single-machine multi-GPU
8
tiled-shared memory
13–14
visualization
6–7
zero-copy host-device I/O
15
Monitored nets, gate-level simulation
358–359
Monte Carlo (MC) algorithm
Chaos Game
265
Fractal Flames
269
radiographic image simulation
algorithms, implementations, evaluations
815–822
basic problem
813–814
code testing
822–824
core methods
814–815
future work
827–828
MC for x-ray tracks
816–818
multiple-GPU CT scan simulations
818–819
performance analysis
822–827
random number generation
820–821
realistic human phantom
824–827
technical details
821–822
x-ray imaging system model
815–816
Monte Carlo photon transport
absorption sampling
254
basic physics
247–249
complete system
256–258
free path sampling
250–254
implementation results
258–259
overview
249–250
parallel random number generation
256
path divergence
249
radiation dose visualization
260–261
scattering direction
254–256
scattering in media
248
MovingPixelMap
560
MPIs, atlas construction
784–785
MRF
Multiclass support vector machine, core method
295
Multidepth test scheme (MDTS), programmable graphics pipeline in CUDA
432–434
Multielectrode arrays (MEAs), temporal data mining
core methodology
212–214
datasets and testbed
222
one thread per occurrence performance
222–224
one thread per occurrence strategy
215–219
serial episode mining
214
two-pass elimination approach
219–222
two-pass elimination performance
224–226
Multi-GPU calculations
ant colony optimization
340
atlas construction
771–790
Carbon-60
17
CT scan simulations
818–819
DFT
147
LDPC
627
molecular electrostatics
57
Monte Carlo photon transport
258
n
-body simulation
131
SPIRiT MRI
724
template-driven agent-based modeling
324
VSM
451–471
Multilabel Graph Cuts, for computer vision
448–450
Multilevel parallel decomposition, MOs
13
Multinomial logistic regression
Multi-objective genetic algorithm (MOGA), OS-SIRT
697
Multi-objective optimization (MOO), OS-SIRT
697–698
Multiple Debye-Hückel (MDH) method
vs
. direct Coulomb summation
47
Multiprocessor memory schematic
372
Multiresolution background modeling, real-time stereo
475–478
Multiresolution stereo matching, process
478–479
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.