#nn — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #nn, aggregated by home.social.
-
Title: P2: hackathon final conf, generative architectres [2024-05-31 Fri]
noise step by step and learn to remove this noise. After
training it can generate images from noise.Now I am going to make demo about my experience in
hackathon that was huge. I will use Emacs ᕙ( •̀ _ •́ )ᕗ
Org mode, TigerVNC and some conference platform that will
allow to share screen and face at the same time. It is a
task from employer.
😶 #dailyreport #AI #neuralnetworks #nn -
Title: P2: hackathon final conf, generative architectres [2024-05-31 Fri]
noise step by step and learn to remove this noise. After
training it can generate images from noise.Now I am going to make demo about my experience in
hackathon that was huge. I will use Emacs ᕙ( •̀ _ •́ )ᕗ
Org mode, TigerVNC and some conference platform that will
allow to share screen and face at the same time. It is a
task from employer.
😶 #dailyreport #AI #neuralnetworks #nn -
Title: P2: hackathon final conf, generative architectres [2024-05-31 Fri]
noise step by step and learn to remove this noise. After
training it can generate images from noise.Now I am going to make demo about my experience in
hackathon that was huge. I will use Emacs ᕙ( •̀ _ •́ )ᕗ
Org mode, TigerVNC and some conference platform that will
allow to share screen and face at the same time. It is a
task from employer.
😶 #dailyreport #AI #neuralnetworks #nn -
Title: P2: hackathon final conf, generative architectres [2024-05-31 Fri]
noise step by step and learn to remove this noise. After
training it can generate images from noise.Now I am going to make demo about my experience in
hackathon that was huge. I will use Emacs ᕙ( •̀ _ •́ )ᕗ
Org mode, TigerVNC and some conference platform that will
allow to share screen and face at the same time. It is a
task from employer.
😶 #dailyreport #AI #neuralnetworks #nn -
Title: P1: hackathon final conf, generative architectres [2024-05-31 Fri]
2) It is impossible to control GNN due to stochastic
nature. Clear data is required, this may be sintetic
data.3) DPO Direct Performance Optimization - training on pairs
good/bad allow to speed up data labeling. https://arxiv.org/pdf/2305.18290Today I have been readling about generative and diffusion
architectures 🤪. In short: DNN is a networks that add #dailyreport #AI #neuralnetworks #nn -
Title: P1: hackathon final conf, generative architectres [2024-05-31 Fri]
2) It is impossible to control GNN due to stochastic
nature. Clear data is required, this may be sintetic
data.3) DPO Direct Performance Optimization - training on pairs
good/bad allow to speed up data labeling. https://arxiv.org/pdf/2305.18290Today I have been readling about generative and diffusion
architectures 🤪. In short: DNN is a networks that add #dailyreport #AI #neuralnetworks #nn -
Title: P1: hackathon final conf, generative architectres [2024-05-31 Fri]
2) It is impossible to control GNN due to stochastic
nature. Clear data is required, this may be sintetic
data.3) DPO Direct Performance Optimization - training on pairs
good/bad allow to speed up data labeling. https://arxiv.org/pdf/2305.18290Today I have been readling about generative and diffusion
architectures 🤪. In short: DNN is a networks that add #dailyreport #AI #neuralnetworks #nn -
Title: P1: hackathon final conf, generative architectres [2024-05-31 Fri]
2) It is impossible to control GNN due to stochastic
nature. Clear data is required, this may be sintetic
data.3) DPO Direct Performance Optimization - training on pairs
good/bad allow to speed up data labeling. https://arxiv.org/pdf/2305.18290Today I have been readling about generative and diffusion
architectures 🤪. In short: DNN is a networks that add #dailyreport #AI #neuralnetworks #nn -
Title: P0: hackathon final conf, generative architectres [2024-05-31 Fri]
Yesterday I have been at final conference ꙭ of hackathon
in which I have been participated recently. Here was
a professor from AIRI Artificial Intelligence Research
Institute. 🤘He told 👄 that:
1) All Generative NN archintectures are generalized into 2
types:
- Transformer architecture - sequence generator ✯
- Diffusion architecture - iterative refinement ✵ #dailyreport #AI #neuralnetworks #nn -
Title: P0: hackathon final conf, generative architectres [2024-05-31 Fri]
Yesterday I have been at final conference ꙭ of hackathon
in which I have been participated recently. Here was
a professor from AIRI Artificial Intelligence Research
Institute. 🤘He told 👄 that:
1) All Generative NN archintectures are generalized into 2
types:
- Transformer architecture - sequence generator ✯
- Diffusion architecture - iterative refinement ✵ #dailyreport #AI #neuralnetworks #nn -
Title: P0: hackathon final conf, generative architectres [2024-05-31 Fri]
Yesterday I have been at final conference ꙭ of hackathon
in which I have been participated recently. Here was
a professor from AIRI Artificial Intelligence Research
Institute. 🤘He told 👄 that:
1) All Generative NN archintectures are generalized into 2
types:
- Transformer architecture - sequence generator ✯
- Diffusion architecture - iterative refinement ✵ #dailyreport #AI #neuralnetworks #nn -
Title: P0: hackathon final conf, generative architectres [2024-05-31 Fri]
Yesterday I have been at final conference ꙭ of hackathon
in which I have been participated recently. Here was
a professor from AIRI Artificial Intelligence Research
Institute. 🤘He told 👄 that:
1) All Generative NN archintectures are generalized into 2
types:
- Transformer architecture - sequence generator ✯
- Diffusion architecture - iterative refinement ✵ #dailyreport #AI #neuralnetworks #nn -
Title: P1: Tensorflow decentralized training [2023-08-04 Fri]
I used class based approach to import model from tfm
and wrap it in another tf.keras.Model.Here is very good code that I used with interesting "Arc Distance" loss
function for images classification.https://www.kaggle.com/code/alifrahman/landmark-recognition2020-google\n#k8s #nn #ai #neural #tensorflow #k8s
-
Title: P1: Tensorflow decentralized training [2023-08-04 Fri]
I used class based approach to import model from tfm
and wrap it in another tf.keras.Model.Here is very good code that I used with interesting "Arc Distance" loss
function for images classification.https://www.kaggle.com/code/alifrahman/landmark-recognition2020-google\n#k8s #nn #ai #neural #tensorflow #k8s
-
Title: P1: Tensorflow decentralized training [2023-08-04 Fri]
I used class based approach to import model from tfm
and wrap it in another tf.keras.Model.Here is very good code that I used with interesting "Arc Distance" loss
function for images classification.https://www.kaggle.com/code/alifrahman/landmark-recognition2020-google\n#k8s #nn #ai #neural #tensorflow #k8s
-
Title: P1: Tensorflow decentralized training [2023-08-04 Fri]
I used class based approach to import model from tfm
and wrap it in another tf.keras.Model.Here is very good code that I used with interesting "Arc Distance" loss
function for images classification.https://www.kaggle.com/code/alifrahman/landmark-recognition2020-google\n#k8s #nn #ai #neural #tensorflow #k8s
-
Title: P0: Tensorflow decentralized training [2023-08-04 Fri]
I was studying TensorFlow Model Garden github.com/tensorflow/models.
There is two approaches to instantiate model:
1) with config factory (Experiment factory). It uses JSON/YAML to configure model
2) Class based approach with tf.keras.ModelI created decentralized training (on #k8s) of RezNet50 with ParameterServer
on landmark 2020 dataset, with help of my small model created in last week.\n#k8s #k8s #nn #ai #neural #tensorflow #k8s -
Title: P0: Tensorflow decentralized training [2023-08-04 Fri]
I was studying TensorFlow Model Garden github.com/tensorflow/models.
There is two approaches to instantiate model:
1) with config factory (Experiment factory). It uses JSON/YAML to configure model
2) Class based approach with tf.keras.ModelI created decentralized training (on #k8s) of RezNet50 with ParameterServer
on landmark 2020 dataset, with help of my small model created in last week.\n#k8s #k8s #nn #ai #neural #tensorflow #k8s -
Title: P0: Tensorflow decentralized training [2023-08-04 Fri]
I was studying TensorFlow Model Garden github.com/tensorflow/models.
There is two approaches to instantiate model:
1) with config factory (Experiment factory). It uses JSON/YAML to configure model
2) Class based approach with tf.keras.ModelI created decentralized training (on #k8s) of RezNet50 with ParameterServer
on landmark 2020 dataset, with help of my small model created in last week.\n#k8s #k8s #nn #ai #neural #tensorflow #k8s -
Title: P0: Tensorflow decentralized training [2023-08-04 Fri]
I was studying TensorFlow Model Garden github.com/tensorflow/models.
There is two approaches to instantiate model:
1) with config factory (Experiment factory). It uses JSON/YAML to configure model
2) Class based approach with tf.keras.ModelI created decentralized training (on #k8s) of RezNet50 with ParameterServer
on landmark 2020 dataset, with help of my small model created in last week.\n#k8s #k8s #nn #ai #neural #tensorflow #k8s -
• "Parallel Distributed Processing", David Rumelhart
• "Analog VLSI and Neural Systems", Carver MeadThese are the two books that nudged me into #NN #AI, way back in the 1980s. These books provide the historical context of, and the foundational knowledge in, connectionist paradigm and neuromorphic computing. Those are sufficient enticements for a #CS with an EE bent, as well as an #EE with a CS leaning, to devour them, or at least skim them with relish.
Through the years, I have recommended these books to my younger #IT colleagues who professed an interest in #AI—and that includes everyone, nowadays. No takers, so far.
Years ago, the kids argued that these books were already too old and, hence, had no relevance to their modern pursuits. Wrong, they were.
The kids today make what they believe to be a stronger argument: they have no need for something so old fashioned as reading.
-
• "Parallel Distributed Processing", David Rumelhart
• "Analog VLSI and Neural Systems", Carver MeadThese are the two books that nudged me into #NN #AI, way back in the 1980s. These books provide the historical context of, and the foundational knowledge in, connectionist paradigm and neuromorphic computing. Those are sufficient enticements for a #CS with an EE bent, as well as an #EE with a CS leaning, to devour them, or at least skim them with relish.
Through the years, I have recommended these books to my younger #IT colleagues who professed an interest in #AI—and that includes everyone, nowadays. No takers, so far.
Years ago, the kids argued that these books were already too old and, hence, had no relevance to their modern pursuits. Wrong, they were.
The kids today make what they believe to be a stronger argument: they have no need for something so old fashioned as reading.
-
• "Parallel Distributed Processing", David Rumelhart
• "Analog VLSI and Neural Systems", Carver MeadThese are the two books that nudged me into #NN #AI, way back in the 1980s. These books provide the historical context of, and the foundational knowledge in, connectionist paradigm and neuromorphic computing. Those are sufficient enticements for a #CS with an EE bent, as well as an #EE with a CS leaning, to devour them, or at least skim them with relish.
Through the years, I have recommended these books to my younger #IT colleagues who professed an interest in #AI—and that includes everyone, nowadays. No takers, so far.
Years ago, the kids argued that these books were already too old and, hence, had no relevance to their modern pursuits. Wrong, they were.
The kids today make what they believe to be a stronger argument: they have no need for something so old fashioned as reading.
-
• "Parallel Distributed Processing", David Rumelhart
• "Analog VLSI and Neural Systems", Carver MeadThese are the two books that nudged me into #NN #AI, way back in the 1980s. These books provide the historical context of, and the foundational knowledge in, connectionist paradigm and neuromorphic computing. Those are sufficient enticements for a #CS with an EE bent, as well as an #EE with a CS leaning, to devour them, or at least skim them with relish.
Through the years, I have recommended these books to my younger #IT colleagues who professed an interest in #AI—and that includes everyone, nowadays. No takers, so far.
Years ago, the kids argued that these books were already too old and, hence, had no relevance to their modern pursuits. Wrong, they were.
The kids today make what they believe to be a stronger argument: they have no need for something so old fashioned as reading.
-
• "Parallel Distributed Processing", David Rumelhart
• "Analog VLSI and Neural Systems", Carver MeadThese are the two books that nudged me into #NN #AI, way back in the 1980s. These books provide the historical context of, and the foundational knowledge in, connectionist paradigm and neuromorphic computing. Those are sufficient enticements for a #CS with an EE bent, as well as an #EE with a CS leaning, to devour them, or at least skim them with relish.
Through the years, I have recommended these books to my younger #IT colleagues who professed an interest in #AI—and that includes everyone, nowadays. No takers, so far.
Years ago, the kids argued that these books were already too old and, hence, had no relevance to their modern pursuits. Wrong, they were.
The kids today make what they believe to be a stronger argument: they have no need for something so old fashioned as reading.
-
How modern technology is providing deeper health insights than ever before https://www.byteseu.com/1792425/ #NN #Technology
-
👀The Neural Network Factory: An LLM-Generated Dataset 👀
We leveraged GPT-5 to automatically generate a diverse dataset of neural networks suitable for empirical experimentation.
The dataset contains 608 neural networks implemented in PyTorch, each defined by explicit design choices across four key dimensions: architecture type, task category, input data characteristics, and model complexity.
-- Work led by Nadia DAOUDI
-
👀The Neural Network Factory: An LLM-Generated Dataset 👀
We leveraged GPT-5 to automatically generate a diverse dataset of neural networks suitable for empirical experimentation.
The dataset contains 608 neural networks implemented in PyTorch, each defined by explicit design choices across four key dimensions: architecture type, task category, input data characteristics, and model complexity.
-- Work led by Nadia DAOUDI
-
👀The Neural Network Factory: An LLM-Generated Dataset 👀
We leveraged GPT-5 to automatically generate a diverse dataset of neural networks suitable for empirical experimentation.
The dataset contains 608 neural networks implemented in PyTorch, each defined by explicit design choices across four key dimensions: architecture type, task category, input data characteristics, and model complexity.
-- Work led by Nadia DAOUDI
-
👀The Neural Network Factory: An LLM-Generated Dataset 👀
We leveraged GPT-5 to automatically generate a diverse dataset of neural networks suitable for empirical experimentation.
The dataset contains 608 neural networks implemented in PyTorch, each defined by explicit design choices across four key dimensions: architecture type, task category, input data characteristics, and model complexity.
-- Work led by Nadia DAOUDI
-
👀The Neural Network Factory: An LLM-Generated Dataset 👀
We leveraged GPT-5 to automatically generate a diverse dataset of neural networks suitable for empirical experimentation.
The dataset contains 608 neural networks implemented in PyTorch, each defined by explicit design choices across four key dimensions: architecture type, task category, input data characteristics, and model complexity.
-- Work led by Nadia DAOUDI
