User profiles for M. Honarkhah

Mehrdad Honarkhah

Data Scientist, Facebook
Verified email at fb.com
Cited by 507

Stochastic simulation of patterns using distance-based pattern modeling

M Honarkhah, J Caers - Mathematical Geosciences, 2010 - Springer
… The details of the algorithm with an experimental proof is provided in Honarkhah and Caers
… A more quantitative comparison is provided in Honarkhah and Caers (2009), where the …

Direct pattern-based simulation of non-stationary geostatistical models

M Honarkhah, J Caers - Mathematical geosciences, 2012 - Springer
Non-stationary models often capture better spatial variation of real world spatial phenomena
than stationary ones. However, the construction of such models can be tedious as it …

AutoDiff: combining Auto-encoder and Diffusion model for tabular data synthesizing

N Suh, X Lin, DY Hsieh, M Honarkhah… - arXiv preprint arXiv …, 2023 - arxiv.org
Diffusion model has become a main paradigm for synthetic data generation in many subfields
of modern machine learning, including computer vision, language model, or speech …

Offshore gas condensate field history-match and predictions: Ensuring probabilistic forecasts are built with diversity in mind

H Gross, M Honarkhah, Y Chen - SPE Asia Pacific Oil and Gas …, 2011 - onepetro.org
Ideally, probabilistic forecasts improve the decision quality, capture upsides and mitigate
downsides. Unfortunately, one of the risks associated with probabilistic forecasts is the possible …

Application of ISO 10001 and ISO 10002 in engineering courses

M Honarkhah - 2010 - era.library.ualberta.ca
An application of ISO 10001 and ISO 10002 customer satisfaction standards to enhance
students’ satisfaction in engineering courses is studied. An ISO 10001-based Code …

Building large machine learning models from small distributed models: A layer matching approach

…, B Song, M Honarkhah, J Ding, M Hong - … on Federated Learning …, 2022 - openreview.net
… Assume that we set the maximum number of layers in block m in the server 278 model
asL0,m. Then the costs for matching all layers in block m for client M form a cost matrix 279 …

[PDF][PDF] Classifying existing and generating new training image patterns in kernel space

M Honarkhah, J Caers - 21st SCRF ALIATE Meeting, Stanford …, 2008 - pangea.stanford.edu
Advent of Multiple-Point Geostatistics helped to integrate complex subsurface geological
structures and features into the model by the concept of training images. Initial algorithms such …

The fundamentals of longevity risk

P Nakada, C Breaux, M Honarkhah… - Special …, 2015 - guides.pm-research.com
The authors’ firm, Risk Management Solutions (RMS), has had a front-row seat in the
development of catastrophe-linked securities from an esoteric, fringe asset class to a mainstream, …

Back-etched super-junction LDMOST on SOI

S Honarkhah, S Nassif-Khalil… - Proceedings of the 30th …, 2004 - ieeexplore.ieee.org
… on a 0.8 /spl mu/m SOI film were investigated. The device with 15.5 /spl mu/m of SJ region
exhibits a breakdown voltage of 317 V, a specific on-resistance of 48.3 m/spl Omega/cm/sup 2…

[PDF][PDF] Direct non-stationary multiple-point modeling by distance-based pattern simulation

M Honarkhah, J Caers - Ninth International Geostatistics …, 2012 - geostats2012.nr.no
Most geostatistical techniques rely on some form of stationarity in their approach to creating
model realizations. This assumption of stationarity is inherent to any statistical modeling …