Zobrazeno 1 - 8
of 8
pro vyhledávání: '"Josh Kalin"'
Autor:
Josh Kalin
Simplify next-generation deep learning by implementing powerful generative models using Python, TensorFlow and KerasKey FeaturesUnderstand the common architecture of different types of GANsTrain, optimize, and deploy GAN applications using TensorFlow
Publikováno v:
AIAA AVIATION 2021 FORUM.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1e96c30f1315da15a429cd2a0f4968ad
https://doi.org/10.2514/6.2021-2958
https://doi.org/10.2514/6.2021-2958
Publikováno v:
Computer Science & Information Technology (CS & IT).
Machine learning models present a risk of adversarial attack when deployed in production. Quantifying the contributing factors and uncertainties using empirical measures could assist the industry with assessing the risk of downloading and deploying c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3dea5201a9dd7ae88593559abac25008
https://doi.org/10.5121/csit.2021.111001
https://doi.org/10.5121/csit.2021.111001
Image classification is a common step in image recognition for machine learning in overhead applications. When applying popular model architectures like MobileNetV2, known vulnerabilities expose the model to counter-attacks, either mislabeling a know
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fd5593213e145c9736111a09c74817d6
http://arxiv.org/abs/2103.15897
http://arxiv.org/abs/2103.15897
Taking advantage of computationally lightweight, but high-quality translators prompt consideration of new applications that address neglected languages. For projects with protected or personal data, translators for less popular or low-resource langua
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1ae2700bdcc7ea6cc0b7435b8f1ba797
http://arxiv.org/abs/2101.01628
http://arxiv.org/abs/2101.01628
Publikováno v:
AI4I
This work applies natural language modeling to generate plausible strategic moves in the ancient game of Go. We train the Generative Pretrained Transformer (GPT-2) to mimic the style of Go champions as archived in Smart Game Format (SGF), which offer
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c921de63f0024a04dcf80a59453d6e0b
https://doi.org/10.1109/ai4i49448.2020.00012
https://doi.org/10.1109/ai4i49448.2020.00012
Publikováno v:
Computer Science & Information Technology.
This work proposes a structured approach to baselining a model, identifying attack vectors, and securing the machine learning models after deployment. This method for securing each model post deployment is called the BAD (Build, Attack, and Defend) A
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4c4bdae0d926a643fadc75e3c01f290c
https://doi.org/10.5121/csit.2020.100609
https://doi.org/10.5121/csit.2020.100609
Publikováno v:
Automatic Target Recognition XXX.
Single Image Super Resolution (SISR) is the process of mapping a low-resolution image to a high resolution image. This inherently has applications in remote sensing as a way to increase the spatial resolution in satellite imagery. This suggests a pos
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1994b9e49c05981ddfc75bcbf4420cae
https://doi.org/10.1117/12.2557845
https://doi.org/10.1117/12.2557845