BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Observifolds: path integral contour deformation
DTSTART;VALUE=DATE-TIME:20210929T130000Z
DTEND;VALUE=DATE-TIME:20210929T135000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2344@indico.ectstar.eu
DESCRIPTION:Speakers: William Detmold\n\nhttps://indico.ectstar.eu/event/7
7/contributions/2344/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2344/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Feature extraction of machine learning and phase transition point
of Ising model
DTSTART;VALUE=DATE-TIME:20210930T080000Z
DTEND;VALUE=DATE-TIME:20210930T085000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2346@indico.ectstar.eu
DESCRIPTION:Speakers: Shotaro Shiba Funai (Okinawa Institute of Science an
d Technology)\n\nhttps://indico.ectstar.eu/event/77/contributions/2346/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2346/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Quantitative analysis of phase transitions in two-dimensional XY m
odels using persistent homology
DTSTART;VALUE=DATE-TIME:20210929T144000Z
DTEND;VALUE=DATE-TIME:20210929T151000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2345@indico.ectstar.eu
DESCRIPTION:Speakers: Nicholas Sale (Swansea University)\n\nIn this talk I
will introduce persistent homology\, a tool from the emerging field of to
pological data analysis\, and demonstrate how it can be used to produce ne
w observables of lattice spin models. In particular\, I will talk about re
cent work on developing a persistent homology-based methodology to extract
the critical temperature and critical exponent of the correlation length
of phase transitions in three variants of the two-dimensional XY model\n\n
https://indico.ectstar.eu/event/77/contributions/2345/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2345/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine learning with quantum field theories
DTSTART;VALUE=DATE-TIME:20210927T144000Z
DTEND;VALUE=DATE-TIME:20210927T151000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2336@indico.ectstar.eu
DESCRIPTION:Speakers: Dimitrios Bachtis (Swansea University (UK))\n\nThe
exact equivalence between lattice field theories and the mathematical fram
ework of Markov random fields opens up the opportunity to investigate mach
ine learning from the perspective of quantum field theory. In this talk w
e prove Markov properties for the $\\phi^{4}$ theory and we then derive $\
\phi^{4}$ neural networks which can be viewed as generalizations of conven
tional neural network architectures. Finally\, applications pertinent to t
he minimization of an asymmetric distance between the probability distribu
tion of the $\\phi^{4}$ machine learning algorithms and target probability
distributions are additionally presented.\n\nhttps://indico.ectstar.eu/ev
ent/77/contributions/2336/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2336/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Critical temperature from unsupervised deep learning autoencoders
DTSTART;VALUE=DATE-TIME:20210928T085000Z
DTEND;VALUE=DATE-TIME:20210928T094000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2337@indico.ectstar.eu
DESCRIPTION:Speakers: Srijit Paul (Johannes Gutenberg University Mainz)\n
\nWe discuss deep learning autoencoders for the unsupervised recognition o
f phase transitions in physical systems formulated on a lattice. We elabor
ate on the applicability and limitations of this deep learning model in te
rms of extracting the relevant physics. The results are shown in context o
f 2D\, 3D and 4D Ising\, phi^4 and XY models.\n\nhttps://indico.ectstar.eu
/event/77/contributions/2337/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2337/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine Learning to learn physics? Trials and questions
DTSTART;VALUE=DATE-TIME:20210927T080000Z
DTEND;VALUE=DATE-TIME:20210927T085000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2332@indico.ectstar.eu
DESCRIPTION:Speakers: Marco Cristoforetti (FBK)\n\nhttps://indico.ectstar.
eu/event/77/contributions/2332/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2332/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine learning for theories with fermions
DTSTART;VALUE=DATE-TIME:20210929T135000Z
DTEND;VALUE=DATE-TIME:20210929T144000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2350@indico.ectstar.eu
DESCRIPTION:Speakers: Andrei Alexandru (The George Washington University)\
n\nMachine learning can be used for generative methods that approximate PD
Fs corresponding to quantum field theories. To remove any bias an accept/r
eject step is required which for theories with fermions involve the calcul
ation of the fermionic determinant. We investigate the use of pseudo-fermi
on methods for the accept/reject step to bypass the need to compute costly
determinants. As an example we use the two dimensional Thirring model.\n\
nhttps://indico.ectstar.eu/event/77/contributions/2350/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2350/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Gauge equivariant neural networks for quantum lattice gauge theori
es
DTSTART;VALUE=DATE-TIME:20210930T144000Z
DTEND;VALUE=DATE-TIME:20210930T151000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2351@indico.ectstar.eu
DESCRIPTION:Speakers: Di Luo (University of illinois\, Urbana-Champaign)\n
\nI will discuss our recent advancement on neural network quantum states w
ith gauge theories for quantum lattice models. I will first introduce the
gauge equivariant neural-network quantum states for quantum lattice gauge
theories with Zd gauge group and non-abelian Kitaev D(G) models. In parti
cular\, the neural network representation is combined with variational qua
ntum Monte Carlo to demonstrate the confining/deconfining phase transition
in Z2 lattice gauge theory. After that I will present another gauge invar
iant autoregressive neural network approach for ground state and real time
simulations in a variety of quantum lattice models.\n\nhttps://indico.ect
star.eu/event/77/contributions/2351/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2351/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Interpreting artificial neural networks in the context of theoreti
cal physics
DTSTART;VALUE=DATE-TIME:20210927T085000Z
DTEND;VALUE=DATE-TIME:20210927T094000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2333@indico.ectstar.eu
DESCRIPTION:Speakers: Sebastian Wetzel (Perimeter Institute for Theoretica
l Physics)\n\nSince many concepts in theoretical physics are well known to
scientists in the form of equations\, it is possible to identify such con
cepts in non-conventional applications of neural networks to physics. In t
his talk\, we examine what is learned by convolutional neural networks\, a
utoencoders or siamese networks in various physical domains. We find that
these networks intrinsically learn physical concepts like order parameters
\, energies\, or other conserved quantities.\n\nhttps://indico.ectstar.eu/
event/77/contributions/2333/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2333/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Universes as Big Data
DTSTART;VALUE=DATE-TIME:20211001T080000Z
DTEND;VALUE=DATE-TIME:20211001T085000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2352@indico.ectstar.eu
DESCRIPTION:Speakers: Yang-Hui He (London Institute\, Royal Institution)\n
\nhttps://indico.ectstar.eu/event/77/contributions/2352/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2352/
END:VEVENT
BEGIN:VEVENT
SUMMARY:End
DTSTART;VALUE=DATE-TIME:20211001T130000Z
DTEND;VALUE=DATE-TIME:20211001T135000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2354@indico.ectstar.eu
DESCRIPTION:https://indico.ectstar.eu/event/77/contributions/2354/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2354/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Neural autoregressive toolbox for many-body physics
DTSTART;VALUE=DATE-TIME:20210927T135000Z
DTEND;VALUE=DATE-TIME:20210927T144000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2335@indico.ectstar.eu
DESCRIPTION:Speakers: Juan Carrasquilla (Vector Institute for Artificial I
ntelligence\, Toronto (CA))\n\nI will discuss our recent work on the use o
f autoregressive neural networks for many-body physics. In particular\, I
will discuss two approaches to represent quantum states using these models
and their applications to the reconstruction of quantum states\, the simu
lation of real-time dynamics as well as the approximation of ground states
of classical and quantum many-body systems.\n\nhttps://indico.ectstar.eu/
event/77/contributions/2335/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2335/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Using Machine Learning to Alleviate the Sign Problem in the Hubbar
d Model
DTSTART;VALUE=DATE-TIME:20210930T130000Z
DTEND;VALUE=DATE-TIME:20210930T135000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2348@indico.ectstar.eu
DESCRIPTION:Speakers: Thomas Luu (Forschungszentrum Jülich/University of
Bonn)\n\nI will discuss how machine learning can be used to alleviate the
sign problem in stochastic simulations of low-D systems. The method we us
e is based off neural network (NN) approximations of Lefschetz thimbles th
at are determined via holomorphic flow. The target Hamiltonian is the Hub
bard model\, but our application can be adapted to other systems. I provi
de results for non-bipartite systems that have intrinsic sign problems reg
ardless of the presence of a chemical potential\, and also for bi-partite
systems with non-zero chemical potential. I also show how the adaption of
a complex-valued NN with appropriate affine layers can greatly simplify t
he calculation of the determinant of the induced Jacobian\, providing scal
ing that is linear in the volume as opposed to volume^3 for standard det J
calculations.\n\nhttps://indico.ectstar.eu/event/77/contributions/2348/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2348/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Why deep networks generalize
DTSTART;VALUE=DATE-TIME:20211001T085000Z
DTEND;VALUE=DATE-TIME:20211001T094000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2353@indico.ectstar.eu
DESCRIPTION:Speakers: Robert de Mello Koch (Huzhou University and Universi
ty of the Witwatersrand)\n\nTraining a deep network involves applying an a
lgorithm which fixes the parameters of the network. The performance of the
trained deep network is evaluated by studying the trained\nnetwork's perf
ormance on unseen test data. The difference between how the network perfor
ms on the training data and on unseen data defines a generalization error.
Networks that perform as well on unseen data as they did on training data
\, have a small generalization error.\n\nWe have definite expectations for
the size of the generalization error\, based essentially on common sense.
If the training data set is much smaller than the number of parameters in
the network\, training can fit any\ndata perfectly\, so that errors and n
oise are captured during training. Typical deeps network applications use
deep networks with hundreds of millions of parameters\, trained using data
sets with tens of thousands of parameters. Clearly then\, we are squarely
in the regime of large generalization errors. Remarkably however\, for ty
pical deep learning applications\, the generalization error is small. This
begs the question: why do deep nets generalize?\n\nIn this talk we develo
p parallels between deep learning and the renormalization group to suggest
why deep networks generalize.\n\nhttps://indico.ectstar.eu/event/77/contr
ibutions/2353/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2353/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Flow-based generative models for ensemble generation
DTSTART;VALUE=DATE-TIME:20210929T080000Z
DTEND;VALUE=DATE-TIME:20210929T085000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2342@indico.ectstar.eu
DESCRIPTION:Speakers: Gurtej Kanwar (University of Bern)\n\nCritical slowi
ng down and topological freezing cause the Monte Carlo cost of lattice QCD
simulations to severely diverge as the lattice regulator is removed. I wi
ll discuss the application of generative flow-based models to Monte Carlo
sampling for lattice field theory as a means of circumventing these issues
\, in particular covering the construction and evaluation of flow-based sa
mplers in proof-of-principle gauge theory applications. Finally\, I discus
s progress towards including the contributions of fermionic degrees of fre
edom in this method.\n\nhttps://indico.ectstar.eu/event/77/contributions/2
342/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2342/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine Learning Algorithms for faster determination of Lattice QC
D Hadron Correlators
DTSTART;VALUE=DATE-TIME:20210928T144000Z
DTEND;VALUE=DATE-TIME:20210928T151000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2341@indico.ectstar.eu
DESCRIPTION:Speakers: Giovanni Pederiva (Michigan State University)\n\nA b
ig portion of Lattice QCD calculations requires the calculation of hadroni
c two-point correlation functions. These can be computationally challengin
g mostly depending on the size of the systems that are simulated and on th
e physical parameters. We present a new procedure that allows for reduced
computational resources to calculate hadronic two-point functions on the l
attice. We apply a variety of machine learning regression algorithms\, to
relate propagators obtained with the BiCGStab linear solver with different
convergence parameters. A mapping between low precision propagator\ndata
to high precision propagators is investigated and an assessment of the sys
tematic uncertainty over the gauge field configuration ensemble of the pro
cedure is discussed. The validity of the method is assessed based on deriv
ed quantities such as effective masses of hadrons\, together with the pote
ntial gain in computer time\, and on the robustness of the results to the
different models that are tested. The method is found to be stable and to
produce results that are comparable with traditional computations while re
quiring significantly less computer time.\n\nhttps://indico.ectstar.eu/eve
nt/77/contributions/2341/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2341/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Higher-order interactions in statistical physics and machine learn
ing
DTSTART;VALUE=DATE-TIME:20210928T130000Z
DTEND;VALUE=DATE-TIME:20210928T135000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2339@indico.ectstar.eu
DESCRIPTION:Speakers: Ava Khamseh (School of Informatics & Higgs Centre fo
r Theoretical Physics\, The University of Edinburgh)\n\nThe problem of inf
erring pairwise and higher-order interactions in complex systems involving
large numbers of interacting variables\, from observational data\, is fun
damental to many fields. Known to the statistical physics community as the
inverse problem\, it has become accessible in recent years due to real an
d simulated big data being generated. In the first part of this talk\, we
discuss extracting interactions from data using a neural network approach\
, namely the Restricted Boltzmann Machine. In the second part\, we discuss
a model-independent and unbiased estimator of symmetric interactions for
any system of binary and categorical variables\, be it magnetic spins\, no
des in a neural network\, or gene networks in biology. The generality of t
his technique is demonstrated analytically and numerically in various exam
ples.\n\nhttps://indico.ectstar.eu/event/77/contributions/2339/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2339/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine Learning Prediction and Compression of Lattice QCD Observa
bles
DTSTART;VALUE=DATE-TIME:20210928T135000Z
DTEND;VALUE=DATE-TIME:20210928T144000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2340@indico.ectstar.eu
DESCRIPTION:Speakers: Boram Yoon (Los Alamos National Laboratory)\n\nIn la
ttice QCD simulations\, a large number of observables are measured on each
Monte Carlo sample of the QCD universe\, called gauge configuration. Sinc
e the measured observables share the same background gauge configuration\,
their statistical fluctuations are correlated with each other\, and analy
zing such correlation is a well-suited problem for machine learning (ML) a
lgorithms. In this talk\, I will present two ML applications to lattice Q
CD problems: (1) prediction of unmeasured-but-computationally-expensive ob
servables from the cheap observables on each gauge configuration\, and (2)
compression of lattice QCD data using D-Wave quantum annealer as an effic
ient binary optimization algorithm. For both applications\, a bias correct
ion algorithm is applied to estimate and correct the systematic error due
to inexact ML predictions and reconstruction.\n\nhttps://indico.ectstar.eu
/event/77/contributions/2340/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2340/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Deep learning and holographic QCD
DTSTART;VALUE=DATE-TIME:20210930T085000Z
DTEND;VALUE=DATE-TIME:20210930T094000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2347@indico.ectstar.eu
DESCRIPTION:Speakers: Koji Hashimoto (Kyoto University\, Physics Departmen
t)\n\nBulk reconstruction in AdS/CFT correspondence is a key idea revealin
g the mechanism of it\, and various methods were\nproposed to solve the in
verse problem. We use deep learning and identify the neural network as the
emergent geometry\,\nto reconstruct the bulk. The lattice QCD data such a
s chiral condensate\, hadron spectra or Wilson loop is used as input\ndata
to reconstruct the emergent geometry of the bulk. The requirement that th
e bulk geometry is a consistent solution of \nan Einstein-dilaton system d
etermines the bulk dilaton potential backwards\, to complete the reconstru
ction program. \nWe demonstrate the determination of the bulk system from
QCD lattice/experiment data\n\nhttps://indico.ectstar.eu/event/77/contribu
tions/2347/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2347/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Differentiable programming for fundamental physics research: statu
s and perspectives
DTSTART;VALUE=DATE-TIME:20210928T080000Z
DTEND;VALUE=DATE-TIME:20210928T085000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2338@indico.ectstar.eu
DESCRIPTION:Speakers: Tommaso Dorigo (INFN\, Padova)\n\nTake the chain rul
e of differential calculus\, model your system with continuous functions\,
add overparametrization and an effective way to navigate stochastically t
hrough the parameter space in search of an extremum of an utility function
\, and you have all it takes to find an optimal solution to even the harde
st optimization problem. Deep learning\, nowadays “differentiable progra
mming”\, is boosting our reach to previously intractable problems. \nI w
ill look at the status of applications of differentiable programming in re
search in particle physics and related areas\, and make a few observations
of where we are heading.\n\nhttps://indico.ectstar.eu/event/77/contributi
ons/2338/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2338/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Training Topological Samplers for Lattice Gauge Theories
DTSTART;VALUE=DATE-TIME:20210930T135000Z
DTEND;VALUE=DATE-TIME:20210930T144000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2349@indico.ectstar.eu
DESCRIPTION:Speakers: Sam Foreman (Argonne National Laboratory)\n\nThe abi
lity to efficiently draw independent configurations from a general density
function is a major computational challenge that has been studied extensi
vely across a variety of scientific disciplines. In particular\, for High
Energy Physics\, the effort required to generate independent gauge field c
onfigurations is known to scale exponentially as we approach physical latt
ice volumes.\n\nWe discuss ongoing developments towards developing a gener
alized version of the Hamiltonian Monte Carlo (HMC) algorithm that efficie
ntly leverages invertible neural network architectures to help combat this
effect\, and demonstrate its success on a two-dimensional U(1) lattice ga
uge theory.\n\nOur implementation is publicly available at https://www.git
hub.com/saforem2/l2hmc-qcd\n\nhttps://indico.ectstar.eu/event/77/contribut
ions/2349/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2349/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine learning phase transitions in a scalable manner on classic
al and quantum processors
DTSTART;VALUE=DATE-TIME:20210927T130000Z
DTEND;VALUE=DATE-TIME:20210927T135000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2334@indico.ectstar.eu
DESCRIPTION:Speakers: Marina Marinkovic (ETH Zurich)\n\nAs the applicatio
ns of machine learning in lattice gauge theories are moving beyond the toy
models\, the parallelization of learning algorithms and alternative appro
aches to their efficient implementation gains in importance. In this talk\
, I will present two possible avenues to speed up the methods with applica
tions to phase transitions classifications. After the discussion of the su
pport vector machine learning model with a focus on its efficient parallel
ization\, we will move the SVM to a quantum circuit and benchmark it using
the Ising model in two dimensions.\n\nhttps://indico.ectstar.eu/event/77/
contributions/2334/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2334/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Machine learning in optical metrology
DTSTART;VALUE=DATE-TIME:20210929T085000Z
DTEND;VALUE=DATE-TIME:20210929T094000Z
DTSTAMP;VALUE=DATE-TIME:20220516T175400Z
UID:indico-contribution-2343@indico.ectstar.eu
DESCRIPTION:Speakers: Barak Bringlotz (Sight Diagnostics\, Israel)\n\nOpti
cal metrology is a technology for high-precision and high-accuracy charact
erization of samples through the measurement of optical images and signals
. An important ingredient in this characterization is the optical modeling
of the data which often involves the solution of the corresponding Maxwel
l equations or approximations thereof. Another technique for modeling opti
cal signals is machine/deep learning based\, and in this talk I will descr
ibe when such data-driven modeling is appropriate. In particular\, I will
describe the use-case of blood diagnostics through optical metrology which
we implement at Sight Diagnostics ® \, and discuss how we overcame few o
f the challenges we faced in our R&D.\n\nhttps://indico.ectstar.eu/event/7
7/contributions/2343/
LOCATION:Aula Renzo Leonardi (ECT* - Trento)
URL:https://indico.ectstar.eu/event/77/contributions/2343/
END:VEVENT
END:VCALENDAR