Seismic Method Estimate velocity from seismic data.pptx
Speeding Up Vectorized Benchmarking of Optimization Algorithms
1. Introduction CI UGPP MODA Conclusion Reference
Speeding Up
Vectorized Benchmarking
of
Optimization Algorithms
Austrian-Slovenian HPC Meeting 2022 – ASHPC22
Seeblickhotel Grundlsee (Austria) & online
May 31 – June 2, 2022
Aleš Zamuda
ales.zamuda@um.si
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 1/ 35
2. Introduction CI UGPP MODA Conclusion Reference
Introduction: Overview (Focus, Use, Scope)
I This contribution focuses on
speeding up of vectorized
benchmarking, which
I includes optimization
algorithms [1].
I These algorithms are used within
Machine Learning (ML) workloads
together with benchmarking
I in order to compute the
performance of instances of such
algorithms
on a whole benchmark [2].
I Such performance measure
provides a more general evaluation
of an algorithm’s applicability,
I i.e. an intelligence generality.
I However, the computational time
for whole benchmark evaluation
extends significantly
I compared to a single instance
evaluation
I and hence speeding up might be
required
I under time-constrained
conditions, especially when e.g.
used for robotic deep sea
underwater missions [2].
Real examples: science and HPC
[1] Zamuda, A., Lloret, E., Optimizing Data-Driven Models for Summarization as Parallel Tasks. Journal of Computational Science 42,
101101 (2020).
[2] Zamuda, A, Brest, J., Self-adaptive control parameters’ randomization frequency and propagations in differential evolution. Swarm and
Evolutionary Computation 25C, 72-99 (2015).
[3] Zamuda, A., Hernández Sosa, J. D., Success history applied to expert system for underwater glider path planning using differential
evolution. Expert Systems with Applications 119, 155-170 (2019).
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 2/ 35
3. Introduction CI UGPP MODA Conclusion Reference
Introduction: Vectorized Benchmarking Opportunities
A closer observation of execution
times for workloads processed in [2] is
provided in Fig. 1, where it is seen
that the execution time (color of the
patches) changes for different
benchmark executions.
Fig. 1: Execution time of full
benchmarks for different instances of
optimization algorithms. Each patch
presents one full benchmark execution
to evaluate an optimization algorithm.
I Therefore, it is useful to consider speeding up of
benchmarking through vectorization of the tasks that a
benchmark is comprised of.
I These include e.g.,
I parallell data cleaning part of an
individual ML tile [1] or
I synchronization between tasks when
executing parallell geospatial processing
[3].
I To enable the possibilities of data cleaning (preprocessing)
as well as geospatial processing in parallell, such
opportunities first need to be found or designed, if none
yet exist for a problem tackled.
I Therefore, this contribution will highlight some experiences
with finding and designing parallell ML pipelines for
vectorization and observe speedup gained from that.
I The speeding up focus will be on optimization
algorithms within such ML pipelines, but some more
future work possibilities will also be provided.
[2] Zamuda, A, Brest, J., Self-adaptive control parameters’ randomization frequency and propagations in differential evolution. Swarm and
Evolutionary Computation 25C, 72-99 (2015).
[3] Zamuda, A., Hernández Sosa, J. D., Success history applied to expert system for underwater glider path planning using differential
evolution. Expert Systems with Applications 119, 155-170 (2019).
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 3/ 35
4. Introduction CI UGPP MODA Conclusion Reference
Background
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 4/ 35
5. Introduction CI UGPP MODA Conclusion Reference
Background on HPC ML Workloads:
Multi-application domain generalized
Computational Intelligence (CI) through HPC
I With the ubiquity of HPC and Cloud Computing,
I connectible using versatile interfaces for their utilisation,
I example successes of these systems is computational
intelligence.
I One of designs for these services is:
I using evolutionary optimization approaches,
I needing much efficiently parallelizable data processing power.
I There are several application domains of this approach:
I generalized numerical functions problems and
I other parallel real world problems, such as
I text processing,
I molecular modelling,
I evolutionary computer vision, and
I robotics.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 5/ 35
6. Introduction CI UGPP MODA Conclusion Reference
CI Algorithm Design: Control Parameters Self-Adaptation
I Through more suitable values of control parameters the search
process exhibits a better convergence (CI — Computational
Intelligence),
I therefore the search converges faster to better solutions,
which survive with greater probability and they create more
offspring and propagate their control parameters
I Recent study with cca. 10 million runs of SPSRDEMMS:
A. Zamuda, J. Brest. Self-adaptive control parameters’
randomization frequency and propagations in differential
evolution. Swarm and Evolutionary Computation, 2015, vol.
25C, pp. 72-99.
DOI 10.1016/j.swevo.2015.10.007.
– SWEVO 2015 RAMONA / SNIP 5.220
I Deployed using arcsub.
I Update: DISH algorithm —
https://doi.org/10.1016/j.swevo.2018.10.013
(Outperforms on CEC 2017 functions.)
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 6/ 35
7. Introduction CI UGPP MODA Conclusion Reference
Existing Challenges: Benchmarks for Known Environments
Inspired by previous computational optimization competitions in continuous settings
that used test functions for optimization application domains:
I single-objective: CEC 2005, 2013, 2014, 2015
I constrained: CEC 2006, CEC 2007, CEC 2010
I multi-modal: CEC 2010, SWEVO 2016
I black-box (target value): BBOB 2009, COCO 2016
I noisy optimization: BBOB 2009
I large-scale: CEC 2008, CEC 2010
I dynamic: CEC 2009, CEC 2014
I real-world: CEC 2011
I computationally expensive: CEC 2013, CEC 2015
I learning-based: CEC 2015
I 100-digit (50% targets): 2019 joined CEC, SEMCCO, GECCO
I multi-objective: CEC 2002, CEC 2007, CEC 2009, CEC 2014
I bi-objective: CEC 2008
I many objective: CEC 2018
Tuning/ranking/hyperheuristics use. → DEs as usual winner algorithms.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 7/ 35
8. Introduction CI UGPP MODA Conclusion Reference
IEEE Congress on Evolutionary Computation (CEC)
Competitions (1/4)
I Storn, Rainer, and Kenneth V. Price. ”Minimizing the Real Functions of
the ICEC’96 Contest by Differential Evolution.”
International Conference on Evolutionary Computation. 1996.
I ...
I CEC 2005 Special Session / Competition on
Evolutionary Real Parameter single objective optimization
I CEC 2006 Special Session / Competition on
Evolutionary Constrained Real Parameter single objective optimization
I CEC 2007 Special Session / Competition on
Performance Assessment of real-parameter MOEAs
I CEC 2008 Special Session / Competition on
large scale single objective global optimization with bound constraints
I CEC 2008 Scale-Invariant Optimisation Competition ”Mountains or Molehills”
I CEC 2009 Special Session / Competition on
Dynamic Optimization (Primarily composition functions were used)
I CEC 2009 Special Session / Competition on
Performance Assessment of real-parameter MOEAs
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 8/ 35
9. Introduction CI UGPP MODA Conclusion Reference
IEEE Congress on Evolutionary Computation (CEC)
Competitions (2/4)
I CEC 2010 Special Session / Competition on
large-scale single objective global optimization with bound constraints
I CEC 2010 Special Session / Competition on
Evolutionary Constrained Real Parameter single objective optimization
I CEC 2010 Special Session on
Niching Introduces novel scalable test problems
I CEC 2011 Competition on Testing Evolutionary Algorithms on Real-world
Numerical Optimization Problems
I CEC 2013 Special Session / Competition on
Real Parameter Single Objective Optimization
I CEC 2014 Special Session / Competition on
Real Parameter Single Objective Optimization
(incorporates expensive functions)
I CEC 2014: Dynamic MOEA Benchmark Problems
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 9/ 35
10. Introduction CI UGPP MODA Conclusion Reference
IEEE Congress on Evolutionary Computation (CEC)
Competitions (3/4)
I CEC 2015 Special Session / Competition on
Real Parameter Single Objective Optimization (incorporates 3 scenarios)
I CEC 2015 Black Box Optimization Competition
I CEC 2015 Dynamic Multi-Objective Optimization
I CEC 2015 Optimization of Big Data
I CEC 2015 Large Scale Global Optimization
I CEC 2015 Bound Constrained Single-Objective Numerical Optimization
I CEC 2015
Optimisation of Problems with Multiple Interdependent Components
I CEC 2015 Niching Methods for Multimodal Optimization
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 10/ 35
11. Introduction CI UGPP MODA Conclusion Reference
IEEE Congress on Evolutionary Computation (CEC)
Competitions (4/4)
I CEC 2016 Special Session / Competition on
Real Parameter Single Objective Optimization (incorporates 4 scenarios)
I CEC 2016 Big Optimization (BigOpt2016)
I CEC 2016 Niching Methods for Multimodal Optimization
I CEC 2016 Special Session Associated with Competition on
Bound Constrained Single Objective Numerical Optimization
I CEC2017 Special Session / Competition on Real Parameter Single
Objective Optimization (also constrained)
I CEC2018 Special Session / Competition on Real Parameter Single
Objective Optimization
I CEC2019 Special Session / Competition on 100-Digit Challenge on Single
Objective Numerical Optimization
I CEC 2020, CEC 2021, 2022, ...: see IEEE Task Force on Benchmarking
(https://cmte.ieee.org/cis-benchmarking/) listings (incl. IEEE
TEVC, MIT EC, MDPI SIs, GitHub, etc.)
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 11/ 35
12. Introduction CI UGPP MODA Conclusion Reference
Some More Benchmark Function Sets
The Genetic and Evolutionary Computation Conference (GECCO):
I GECCO 2014 Windfarm Layout Optimization Competition
I GECCO 2014 Permutation-based Combinatorial Optimization Problems
I GECCO 2015 Black Box Optimization (BBComp)
I GECCO 2015 Combinatorial Black-Box Optimization (CBBOC)
I GECCO 2019 Workshop & Competition on Numerical Optimization
I GECCO 2022 (Boston) Open Optimization Competition 2022: Better
Benchmarking of Sampling-Based Optimization Algorithms
I GECCO 2022 (Boston) SpOC: Space Optimisation Competition
I Swarm and Evolutionary Computation: Novel benchmark functions for
continuous multimodal optimization with comparative results (2015)
I Benchmarks for natural architecture design:
Spatial Tree Morphology Reconstruction
(seeded/pre-processed/vectorized/multi-objective)
I Benchmarks for ocean exploration with underwater robots: Underwater
Glider Path Planning (unconstraint/constraint/variable/multi-objective)
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 12/ 35
13. Introduction CI UGPP MODA Conclusion Reference
Real World Industry Challenges: Motivation
I Optimization of Real World Industry Challenges (RWIC),
I selected as CEC 2011 Real World Optimization Problems,
I a benchmark set contains functions modelling the problems,
I assessment on all 22 functions of CEC 2011 set,
I functions with constraints are handled with additional care.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 13/ 35
14. Introduction CI UGPP MODA Conclusion Reference
Selected Challenges in the CEC 2011 Benchmark (1/2)
I Decomposition of radio FM signals,
I determination of ternary protein structure
I Lennard-Jones inter-atom energy potential,
I parameterization of a chemical process,
I methylcyclopentane → benzene,
I control parameterization for chemical reaction in a continuous
stirred tank reactor,
I inter-atom potential in covalent Silicon systems
I Tersoff energy potential,
I radar spectrum signal broadcast parameterization,
I spread spectrum radar polly phase code design,
I electrical transmission network expansion planning,
I new lines for transmission selection.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 14/ 35
15. Introduction CI UGPP MODA Conclusion Reference
Selected Challenges in the CEC 2011 Benchmark (2/2)
I Large scale transmission pricing,
I circular antenna array design,
I dynamic economic dispatch (with power generator control),
I static economic load dispatch (of power generated),
I hydrothermal scheduling (among hydro/thermal units),
I spacecraft trajectory optimization,
I Mercury (Messenger),
I Saturn (Cassini).
I The collection includes 22 functions, with constraints as:
1. non-feasible evaluation is NaN when constraints are not met, or
2. the constraints are included in the function evaluation value.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 15/ 35
16. Introduction CI UGPP MODA Conclusion Reference
HPC Application 2:
Underwater Glider: Autonomous, Unmanned, Robotic
I underwater glider – navigating sea oceans,
I Autonomous Underwater Vehicle (AUV)
6=
Unmanned Aerial Vehicle (UAV)
I AUV Slocum model (expertise in domain of ULPGC, work
with J. D. Hernández Sosa)
Images:
”Photo: Richard Watt/MOD” (License: OGL v1.0)
Slocum-Glider-Auvpicture 5.jpg (License: Public Domain)
MiniU.jpg (License: CC-BY-SA 3.0)
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 16/ 35
17. Introduction CI UGPP MODA Conclusion Reference
The Buoyancy Drive and Submarine Probes Usefulness
I Driving ”yoyo” uses little energy, most only on descent and
rise (pump); also for maintaining direction little power is
consumed.
+ Use: improving ocean models with real data,
+ the real data at the point of capture,
+ sampling flow of oil discharges,
+ monitoring cable lines, and
+ real-time monitoring of different
sensor data.
1
http://spectrum.ieee.org/image/1523708
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 17/ 35
18. Introduction CI UGPP MODA Conclusion Reference
Preparations – Simulation Scenarios
https://www.google.si/maps/@28.059806,-15.998355,650054m/data=!3m1!1e3
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 18/ 35
19. Introduction CI UGPP MODA Conclusion Reference
Trajectory Optimization: P201,ESTOC2013 3
+ BigData, MyOcean IBI,
satelite link, GPS location
The real trajectory and collected data is available in a Google Earth KML file at the EGO network:
http://www.ego-network.org/dokuwiki/doku.php?glider=P201,ESTOC2013_3
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 19/ 35
20. Introduction CI UGPP MODA Conclusion Reference
Constrained Differential Evolution Optimization
for Underwater Glider Path Planning
in Sub-mesoscale Eddy Sampling
I Corridor-constrained optimization:
eddy border region sampling
I new challenge for UGPP & DE
I Feasible path area is constrained
I trajectory in corridor around the
border of an ocean eddy
The objective of the glider here is to
sample the oceanographic variables more
efficiently,
while keeping a bounded trajectory
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 20/ 35
21. Introduction CI UGPP MODA Conclusion Reference
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 21/ 35
22. Introduction CI UGPP MODA Conclusion Reference
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 22/ 35
23. Introduction CI UGPP MODA Conclusion Reference
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 23/ 35
24. Introduction CI UGPP MODA Conclusion Reference
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 24/ 35
25. Introduction CI UGPP MODA Conclusion Reference
HoP — New Trajectories:
Success history applied to expert system for underwater
glider path planning using differential evolution
I Improved underwater glider path
planning mission scenarios:
optimization with L-SHADE.
I Several configured algorithms
are also compared to, analysed,
and further improved.
I Outranked all other previous
results from literature and
ranked first in comparison.
I New algorithm yielded
practically stable and
competitive output trajectories.
I UGPP unconstrained scenarios —
contributed significantly to the capacity of
the decision-makers for mission plannings.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 25/ 35
26. Introduction CI UGPP MODA Conclusion Reference
Ranking UGPP —
Benchmarking
Aggregation
I Statistically,
all results
from previous paper
were outperformed.
I Main reasons:
tuning (NP),
parameter control
(L-SHADE).
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 26/ 35
27. Introduction CI UGPP MODA Conclusion Reference
Results & Impact
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 27/ 35
28. Introduction CI UGPP MODA Conclusion Reference
Data Analytics of the Operational Data Monitoring
(MODA)
I MODA: data obtained from individual’s task gridlog/diag
WallTime.
f o r i i n *; do [ −f ” $i / g r i d l o g / diag ” ] | | continue ;
echo −n $( cat $i /CONFIG . sh | xargs )” ”
grep WallTime $i / g r i d l o g / diag
done 2>/dev / n u l l |
sed −e ’ s / export //g ’ −e ’ s/=/ /g ’ |
grep WallTime | grep tauF |
sed −e ’ s / tauF 0 . . . tauCR 0 . . . t / t /g ’ |
sed −e ’ s /s$//g ’ | s o r t −g
I MODA: data obtained from individual’s task gridlog/diag
WallTime.
gnuplot <EOS
s e t t i t l e ”MODA( Wall Time [ i n seconds ] ) of SPSRDEMMS f o r d i f f e r e n t F , CR
( both between 0.0 and 1.0)”
s e t p a l e t t e d e f i n e d (0 0 0 0.5 , 1 0 0 1 , 2 0 0.5 1 , 3 0 1 1 , 4 0.5 1 0.5 ,
5 1 1 0 , 6 1 0.5 0 , 7 1 0 0 , 8 0.5 0 0)
unset z t i c s
s e t y t i c s
s e t x t i c s
s p l o t ’moda ’ u : 2 : 4 : 6 with p o i n t s pt 5 ps 9 p a l e t t e t ””
EOS
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 28/ 35
29. Introduction CI UGPP MODA Conclusion Reference
Data Analytics of the Operational Data Monitoring
(MODA): Example Results
τF τCR Time [s]
0.70 0.12 6167
0.36 0.18 13904
0.10 0.25 15426
0.25 0.90 15643
0.20 0.05 16039
0.05 0.85 16128
0.05 0.05 16586
0.05 1.00 16902
0.15 0.65 17207
0.05 0.60 17269
... ... ...
0.2 0.6 183938
0.95 0.60 189802
0.72 0.18 259200
0.77 0.18 259203
0.77 0.12 259211
0.73 0.14 259214
0.79 0.16 259217
0.78 0.15 259222
0.78 0.19 259226
0.73 0.12 259229
Total combined WallTime time: 1.03658e+08
seconds (≈ 3.2 years).
In practice, when Vectorized: ≈ 1 month on
HPC (jobs 21 Aug 2015 – 24 Sep 2015 ≈ 35
days).
I MODA: data obtained from
individual’s task gridlog/diag
WallTime.
I ML task MODA plotted.
I Speedup: 35 days vs. 3.2 years.
I Capability gained: paper
revision.
I Gain impact: reference.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 29/ 35
30. Introduction CI UGPP MODA Conclusion Reference
Conclusion
Summary: vectorized benchmarking, speed up, and impact.
Thanks!
Acknowledgement: this work is supported by ARRS programme
P2-0041; and DAPHNE, funded by the European Union’s Horizon 2020
research and innovation programme under grant agreement No 957407.
Questions?
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 30/ 35
31. Introduction CI UGPP MODA Conclusion Reference
Biography and References: Organizations
I Associate Professor at University of Maribor, Slovenia
I Continuous research programme funded by Slovenian Research Agency,
P2-0041: Computer Systems, Methodologies, and Intelligent Services
I Associate Editor: Swarm and Evolutionary Computation
I IEEE (Institute of Electrical and Electronics Engineers) senior
I IEEE Computational Intelligence Society (CIS), senior member
I IEEE CIS Task Force on Benchmarking, chair Website link
I IEEE CIS, Slovenia Section Chapter (CH08873), chair
I IEEE Slovenia Section, 2018–2021 vice chair
I IEEE Young Professionals Slovenia, past chair
I ACM SIGEVO (Special Interest Group on Genetic and Evolutionary Computation); EurAI; SLAIS
I Co-operation in Science and Techology (COST) Association Management
Committee, member:
I CA COST Action CA15140: Improving Applicability of Nature-Inspired
Optimisation by Joining Theory and Practice (ImAppNIO), WG3 VC
I ICT COST Action IC1406 High-Performance Modelling and Simulation
for Big Data Applications (cHiPSet); SI-HPC; HPC-RIVR user
I EU H2020 Research and Innovation project, holder for UM part: Integrated
Data Analysis Pipelines for Large-Scale Data Management, HPC, and
Machine Learning (DAPHNE),
https://cordis.europa.eu/project/id/957407
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 31/ 35
32. Introduction CI UGPP MODA Conclusion Reference
Biography and References: Top Publications
I Aleš Zamuda, Elena Lloret. Optimizing Data-Driven Models for Summarization as Parallel Tasks.
Journal of Computational Science, 2020, vol. 42, pp. 101101. DOI 10.1016/j.jocs.2020.101101.
I A. Zamuda, J. D. Hernández Sosa. Success history applied to expert system for underwater glider path
planning using differential evolution. Expert Systems with Applications, 2019, vol. 119, pp. 155-170.
DOI 10.1016/j.eswa.2018.10.048
I A. Viktorin, R. Senkerik, M. Pluhacek, T. Kadavy, A. Zamuda. Distance Based Parameter Adaptation
for Success-History based Differential Evolution. Swarm and Evolutionary Computation, 2019, vol. 50,
pp. 100462. DOI 10.1016/j.swevo.2018.10.013.
I A. Zamuda, J. Brest. Self-adaptive control parameters’ randomization frequency and propagations in
differential evolution. Swarm and Evolutionary Computation, 2015, vol. 25C, pp. 72-99.
DOI 10.1016/j.swevo.2015.10.007.
I A. Zamuda, J. D. Hernández Sosa, L. Adler. Constrained Differential Evolution Optimization for
Underwater Glider Path Planning in Sub-mesoscale Eddy Sampling. Applied Soft Computing, 2016,
vol. 42, pp. 93-118. DOI 10.1016/j.asoc.2016.01.038.
I A. Zamuda, J. D. Hernández Sosa. Differential Evolution and Underwater Glider Path Planning
Applied to the Short-Term Opportunistic Sampling of Dynamic Mesoscale Ocean Structures.
Applied Soft Computing, vol. 24, November 2014, pp. 95-108. DOI 10.1016/j.asoc.2014.06.048.
I A. Zamuda, J. Brest. Vectorized Procedural Models for Animated Trees Reconstruction using
Differential Evolution. Information Sciences, vol. 278, pp. 1-21, 2014. DOI 10.1016/j.ins.2014.04.037.
I A. Zamuda, J. Brest. Environmental Framework to Visualize Emergent Artificial Forest Ecosystems.
Information Sciences, vol. 220, pp. 522-540, 2013. DOI 10.1016/j.ins.2012.07.031.
I A. Glotić, A. Zamuda. Short-term combined economic and emission hydrothermal optimization by
surrogate differential evolution. Applied Energy, 1 March 2015, vol. 141, pp. 42-56.
DOI 10.1016/j.apenergy.2014.12.020.
I H. Hamann, Y. Khaluf, J. Botev, M. Divband Soorati, E. Ferrante, O. Kosak, J.-M. Montanier, S.
Mostaghim, R. Redpath, J. Timmis, F. Veenstra, M. Wahby and A. Zamuda. Hybrid Societies:
Challenges and Perspectives in the Design of Collective Behavior in Self-organizing Systems.
Frontiers in Robotics and AI, 2016, vol. 3, no. 14. DOI 10.3389/frobt.2016.00014.
I J. Šilc, A. Zamuda. Special Issue on ”Bioinspired Optimization” (guest editors). Informatica - An
International Journal of Computing and Informatics, 2015, vol. 39, no. 2, pp. 1-122.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 32/ 35
33. Introduction CI UGPP MODA Conclusion Reference
Biography and References: Bound Specific to HPC
PROJECTS:
I DAPHNE: Integrated Data Analysis Pipelines for Large-Scale Data
Management, HPC, and Machine Learning
I ICT COST Action IC1406 High-Performance Modelling and Simulation for Big
Data Applications
I SLING: Slovenian national supercomputing network
I SI-HPC: Slovenian corsortium for High-Performance Computing
I UM HPC-RIVR: Supercomputer at UM, https://www.hpc-rivr.si/
I SmartVillages: Smart digital transformation of villages in the Alpine Space
I Interreg Alpine Space,
https://www.alpine-space.eu/projects/smartvillages/en/home
I Interactive multimedia digital signage (PKP, Adin DS)
EDITOR:
I SWEVO (Top Journal), Associate Editor
I Mathematics-MDPI, Special Issue: Evolutionary Algorithms in Engineering Design Optimization
I Journal of advanced engineering and computation (member of editorial board since 2019). Viet Nam: Ton
Duc Thang University, 2017-. ISSN 2588-123X.
I Cloud Computing and Data Science (Associate Editor, since 2019). Universal Wiser Publisher Pte.Ltd.
I D. Gleich, P. Planinšič, A. Zamuda. 2018 25th International Conference on Systems, Signals and Image
Processing (IWSSIP). IEEE Xplore, Maribor, 20-22 June 2018.
I General Chair: 7-th Joint International Conferences on Swarm, Evolutionary and Memetic Computing
Conference (SEMCCO 2019) & Fuzzy And Neural Computing Conference (FANCCO 2019), Maribor,
Slovenia, EU, 10-12 July 2019; co-editors: Aleš Zamuda, Swagatam Das, Ponnuthurai Nagaratnam
Suganthan, Bijaya Ketan Panigrahi.
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 33/ 35
34. Introduction CI UGPP MODA Conclusion Reference
Biography and References: More on HPC
RESEARCH PUBLICATIONS:
I Aleš Zamuda, Elena Lloret. Optimizing Data-Driven Models for Summarization as Parallel Tasks.
Journal of Computational Science, 2020, vol. 42, pp. 101101. DOI 10.1016/j.jocs.2020.101101.
I Aleš Zamuda, Vincenzo Crescimanna, Juan C. Burguillo, Joana Matos Dias, Katarzyna Wegrzyn-Wolska,
Imen Rached, Horacio González-Vélez, Roman Senkerik, Claudia Pop, Tudor Cioara, Ioan Salomie, Andrea
Bracciali. Forecasting Cryptocurrency Value by Sentiment Analysis: An HPC-Oriented Survey of the
State-of-the-Art in the Cloud Era. Kolodziej J., González-Vélez H. (eds) High-Performance Modelling and
Simulation for Big Data Applications. Lecture Notes in Computer Science, vol 11400, 2019, pp. 325-349.
DOI 10.1007/978-3-030-16272-6 12.
I Nenad Karolija, Aleš Zamuda. On cloud-supported web-based integrated development environment for
programming dataflow architectures. MILUTINOVIĆ, Veljko (ur.), KOTLAR, Milos (ur.). Exploring the
DataFlow supercomputing paradigm: example algorithms for selected applications, (Computer
communications and networks (Internet), ISSN 2197-8433), 2019, pp. 41-51. DOI
10.1007/978-3-030-13803-5 2.
I Simone Spolaor, Marco Gribaudo, Mauro Iacono, Tomas Kadavy, Zuzana Komı́nková Oplatková, Giancarlo
Mauri, Sabri Pllana, Roman Senkerik, Natalija Stojanovic, Esko Turunen, Adam Viktorin, Salvatore
Vitabile, Aleš Zamuda, Marco S. Nobile. Towards Human Cell Simulation. Kolodziej J., González-Vélez H.
(eds) High-Performance Modelling and Simulation for Big Data Applications. Lecture Notes in
Computer Science, vol 11400, 2019, pp. 221-249. DOI 10.1007/978-3-030-16272-6 8.
I A. Zamuda, J. D. Hernandez Sosa, L. Adler. Improving Constrained Glider Trajectories for Ocean Eddy
Border Sampling within Extended Mission Planning Time. IEEE Congress on Evolutionary Computation
(CEC) 2016, 2016, pp. 1727-1734.
I A. Zamuda. Function evaluations upto 1e+12 and large population sizes assessed in distance-based
success history differential evolution for 100-digit challenge and numerical optimization scenarios
(DISHchain 1e+12): a competition entry for ”100-digit challenge, and four other numerical optimization
competitions” at the genetic and evolutionary computation conference (GECCO) 2019. Proceedings of the
Genetic and Evolutionary Computation Conference Companion (GECCO 2019), 2019, pp. 11-12.
I ... several more experiments for papers run using HPCs.
I ... also, pedagogic materials in Slovenian and English — see Conclusion .
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 34/ 35
35. Introduction CI UGPP MODA Conclusion Reference
Promo materials: Calls for Papers, Informational Websites
CS FERI WWW
CIS TFoB
CFPs WWW
LinkedIn
Twitter
Aleš Zamuda 7@aleszamuda Speeding Up Vectorized Benchmarking of Optimization Algorithms 35/ 35