Learning to Learn by Gradient Descent by Gradient DescentKaty Lee
This document discusses learning to learn by training a neural network (LSTM) to be an optimizer that learns update rules rather than using hand-designed update rules. The optimizer takes gradients as input and outputs updates to the parameters of the optimizee network. The optimizer is trained end-to-end using its trajectory optimization objective. Experiments show the learned optimizer can generalize to different network architectures but not different activation functions. The conclusion suggests emailing authors if confused by typos.
Classifying Multi-Variate Time Series at Scale:
Characterizing and understanding the runtime behavior of large scale Big Data production systems is extremely important. Typical systems consist of hundreds to thousands of machines in a cluster with hundreds of terabytes of storage costing millions of dollars, solving problems that are business critical. By instrumenting each running process, and measuring their resource utilization including CPU, Memory, I/O, network etc., as time series it is possible to understand and characterize the workload on these massive clusters. Each time series is a series consisting of tens to tens of thousands of data points that must be ingested and then classified. At Pepperdata, our instrumentation of the clusters collects over three hundred metrics from each task every five seconds resulting in millions of data points per hour. At this scale the data are equivalent to the biggest IOT data sets in the world. Our objective is to classify the collection of time series into a set of classes that represent different work load types. Or phrased differently, our problem is essentially the problem of classifying multivariate time series.
In this talk, we propose a unique, off-the-shelf approach to classifying time series that achieves near best-in-class accuracy for univariate series and generalizes to multivariate time series. Our technique maps each time series to a Grammian Angular Difference Field (GADF), interprets that as an image, uses Google’s pre-trained CNN (trained on Inception v3) to map the GADF images into a 2048-dimensional vector space and then uses a small MLP with two hidden layers, with fifty nodes in each layer, and a softmax output to achieve the final classification. Our work is not domain specific – a fact proven by our achieving competitive accuracies with published results on the univariate UCR data set as well as the multivariate UCI data set.
Bio: Before joining Pepperdata, Ash was executive chairman for Marianas Labs, a deep learning startup sold in December 2015. Prior to that he was CEO for Graphite Systems, a big data storage startup that was sold to EMC DSSD in August 2015. Munshi also served as CTO of Yahoo, as a CEO of both public and private companies, and is on the board of several technology startups.
[딥논읽] Meta-Transfer Learning for Zero-Shot Super-Resolution paper reviewtaeseon ryu
105번째 논문리뷰,
오늘 소개 드릴 논문은 2020 CVPR에서 발표된 Meta-Transfer Learning for Zero-Shot Super-Resolution 라는 논문입니다!
제목에서 유추가 가능하신것 처럼 학습 데이터없이 저해상도 사진을 고해상도 사진으로 바꿔주는 Zero Shot Super Resolution을 위한 Meta Transfer Learning을 소개합니다. Internal Learning에 적합한 General한 초기 parameter를 찾는것에 기반하여 한번의 Gradient Update만으로 최적의 성능을 보여주는것 방법에 대해서 소개합니다.
논문에 대한 자세한 리뷰를 이미지 처리팀 김선옥 님이 자세한 리뷰 도와주셨습니다!
https://youtu.be/lEqbXLrUlW4
Presentation in Vietnam Japan AI Community in 2019-05-26.
The presentation summarizes what I've learned about Regularization in Deep Learning.
Disclaimer: The presentation is given in a community event, so it wasn't thoroughly reviewed or revised.
Learning to Learn by Gradient Descent by Gradient DescentKaty Lee
This document discusses learning to learn by training a neural network (LSTM) to be an optimizer that learns update rules rather than using hand-designed update rules. The optimizer takes gradients as input and outputs updates to the parameters of the optimizee network. The optimizer is trained end-to-end using its trajectory optimization objective. Experiments show the learned optimizer can generalize to different network architectures but not different activation functions. The conclusion suggests emailing authors if confused by typos.
Classifying Multi-Variate Time Series at Scale:
Characterizing and understanding the runtime behavior of large scale Big Data production systems is extremely important. Typical systems consist of hundreds to thousands of machines in a cluster with hundreds of terabytes of storage costing millions of dollars, solving problems that are business critical. By instrumenting each running process, and measuring their resource utilization including CPU, Memory, I/O, network etc., as time series it is possible to understand and characterize the workload on these massive clusters. Each time series is a series consisting of tens to tens of thousands of data points that must be ingested and then classified. At Pepperdata, our instrumentation of the clusters collects over three hundred metrics from each task every five seconds resulting in millions of data points per hour. At this scale the data are equivalent to the biggest IOT data sets in the world. Our objective is to classify the collection of time series into a set of classes that represent different work load types. Or phrased differently, our problem is essentially the problem of classifying multivariate time series.
In this talk, we propose a unique, off-the-shelf approach to classifying time series that achieves near best-in-class accuracy for univariate series and generalizes to multivariate time series. Our technique maps each time series to a Grammian Angular Difference Field (GADF), interprets that as an image, uses Google’s pre-trained CNN (trained on Inception v3) to map the GADF images into a 2048-dimensional vector space and then uses a small MLP with two hidden layers, with fifty nodes in each layer, and a softmax output to achieve the final classification. Our work is not domain specific – a fact proven by our achieving competitive accuracies with published results on the univariate UCR data set as well as the multivariate UCI data set.
Bio: Before joining Pepperdata, Ash was executive chairman for Marianas Labs, a deep learning startup sold in December 2015. Prior to that he was CEO for Graphite Systems, a big data storage startup that was sold to EMC DSSD in August 2015. Munshi also served as CTO of Yahoo, as a CEO of both public and private companies, and is on the board of several technology startups.
[딥논읽] Meta-Transfer Learning for Zero-Shot Super-Resolution paper reviewtaeseon ryu
105번째 논문리뷰,
오늘 소개 드릴 논문은 2020 CVPR에서 발표된 Meta-Transfer Learning for Zero-Shot Super-Resolution 라는 논문입니다!
제목에서 유추가 가능하신것 처럼 학습 데이터없이 저해상도 사진을 고해상도 사진으로 바꿔주는 Zero Shot Super Resolution을 위한 Meta Transfer Learning을 소개합니다. Internal Learning에 적합한 General한 초기 parameter를 찾는것에 기반하여 한번의 Gradient Update만으로 최적의 성능을 보여주는것 방법에 대해서 소개합니다.
논문에 대한 자세한 리뷰를 이미지 처리팀 김선옥 님이 자세한 리뷰 도와주셨습니다!
https://youtu.be/lEqbXLrUlW4
Presentation in Vietnam Japan AI Community in 2019-05-26.
The presentation summarizes what I've learned about Regularization in Deep Learning.
Disclaimer: The presentation is given in a community event, so it wasn't thoroughly reviewed or revised.
○ 개요
현재 많은 연구자들이 network를 깊고 넓게 설계함으로써 높은 인식률을 갖는 네트워크를 얻고 있다. Network의 크기가 증가하면서 parameter와 computation의 수가 증가하게 되었고, 이러한 문제를 해결하기 위하여 pruning을 기반으로 한 압축 알고리즘들이 제안되어 왔다. 하지만 이러한 방법을 이용하여서는 network architecture자체를 바꿀 수 없기 때문에, 구조에서 오는 한계점들은 해결할 수 없었다.
Network recasting은 구조의 특성으로 인하여 발생하는 한계들을 해결하기 위하여 network architecture 자체를 바꾸는 방법이다. Network recasting을 이용하면 network를 구성하고있는 block들을 다른 형태의 block으로 변환을 할 수 있게 된다. Block-wise recasting 방법을 사용하여 각 block들을 변환할 수 있고, 해당 방법을 연속하여 적용함으로써 전체 network의 구조를 바꿀 수 있다. Sequential recasting 방법을 이용하게 되면 inference accuracy를 더욱 잘 보존할 수 있고, 또한 network architecture에 상관 없이 vanishing gradient problem을 완화 시킬 수 있다. Network recasting을 같은 network architecture에 적용하게 되면 parameter와 computation을 줄이는 효과를 얻을 수 있고, 다른 종류의 network architecture로 변환하게 되면 network를 가속시킬 수 있다. 이러한 경우에는 network architecture 자체를 변경할 수 있기 때문에 구조적 한계보다 더 높은 속도 향상을 얻을 수 있다.
Deep Learning Fast MRI Using Channel Attention in Magnitude DomainJoonhyung Lee
My presentation on how we participated in the fastMRI Challanege in 2019.
Aside from theoretical considerations, it also explains key implementation issues that arise in all deep learning for MRI such as disk I/O and CPU/GPU load balancing.
Used for presentation at ISBI 2020 Oral session.
Accidentally wrote the title as "Deep Learning Sum-of-Squares Images in Accelerated Parallel MRI". Sorry for the mistake!
YOLOv4: optimal speed and accuracy of object detection reviewLEE HOSEONG
YOLOv4 builds upon previous YOLO models and introduces techniques like CSPDarknet53, SPP, PAN, Mosaic data augmentation, and modifications to existing methods to achieve state-of-the-art object detection speed and accuracy while being trainable on a single GPU. Experiments show that combining these techniques through a "bag of freebies" and "bag of specials" approach improves classifier and detector performance over baselines on standard datasets. The paper contributes an efficient object detection model suitable for production use with limited resources.
Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017MLconf
This document discusses deep reinforcement learning and concept network reinforcement learning. It begins with an introduction to reinforcement learning concepts like Markov decision processes and value-based methods. It then describes Concept-Network Reinforcement Learning which decomposes complex tasks into high-level concepts or actions. This allows composing existing solutions to sub-problems without retraining. The document provides examples of using concept networks for lunar lander and robot pick-and-place tasks. It concludes by discussing how concept networks can improve sample efficiency, especially for sparse reward problems.
Restricting the Flow: Information Bottlenecks for Attributiontaeseon ryu
101번째 영상,
펀디멘탈팀 김준호 님의
Restricting the Flow: Information Bottlenecks for Attribution
논문 리뷰 입니다
Explanable ai, xai와 관련된 페이퍼 입니다! 관련되어 관심있으신 분들이 많은 도움이 되시길 바랍니다! attribution map을 이용하여 결과물에 영향을 준 네트워크의 gradient를 직접 추적하여 비주얼 explanation을 추적하는 방식입니다! 펀디멘탈팀 김준호님이 밑바닥부터 자세한 리뷰를 도와주셨습니다!
오늘도 많은 관심과 사랑 감사합니다!
PR-343: Semi-Supervised Semantic Segmentation with Cross Pseudo SupervisionSungchul Kim
This document proposes a new semi-supervised learning method called Cross Pseudo Supervision (CPS) for semantic segmentation. CPS trains two segmentation networks simultaneously where each network generates pseudo labels for the other using its own predictions. It incorporates this cross pseudo supervision loss with a CutMix data augmentation technique. The document evaluates CPS on PASCAL VOC 2012 and Cityscapes datasets with limited labeled data, finding it outperforms other semi-supervised segmentation methods and ablations. Qualitative results show CPS with CutMix generates more accurate segmentations than alternatives.
Human uncertainty makes classification more robust, ICCV 2019 ReviewLEE HOSEONG
1. The document summarizes a research paper that proposes training deep neural networks on soft labels representing human uncertainty in image classification, which improves generalization and robustness compared to training on hard labels.
2. Experiments show that models trained on soft labels constructed from human responses better fit patterns of human uncertainty and improve accuracy, cross-entropy, and a new second-best accuracy measure on various generalization datasets.
3. Alternative soft label methods are also explored, finding that human uncertainty provides a more important contribution than soft labels alone. While robustness to adversarial attacks is improved, defenses are still needed.
201907 AutoML and Neural Architecture SearchDaeJin Kim
Brief introduction of NAS
Review of EfficientNet (Google Brain), RandWire (FAIR) papers
NAS flow slide from KihoSuh's slideshare (https://www.slideshare.net/KihoSuh/neural-architecture-search-with-reinforcement-learning-76883153)
[References]
[1] EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (https://arxiv.org/abs/1905.11946)
[2] Exploring Randomly Wired Neural Networks for Image Recognition (https://arxiv.org/abs/1904.01569)
The document presents research on using neural networks to predict Earth Orientation Parameters (EOP) such as UT1-TAI. Three neural network models were tested:
1) Network 1 varied the number of neurons proportionally with increasing training sample size.
2) Network 2 kept the number of neurons constant while increasing sample size.
3) Network 3 used daily training data with 2 neurons and sample sizes of 4, 10, 20, and 365 days.
The goal was to minimize prediction error (RMSE) for horizons of 5-25 days by adjusting sample size and neurons. Results showed the best balance was needed between these factors, and that short-term prediction was possible within 10 days using
Dueling Network Architectures for Deep Reinforcement LearningYoonho Lee
This document summarizes reinforcement learning algorithms like Deep Q-Network (DQN), Double DQN, prioritized experience replay, and the Dueling Network architecture. DQN uses a deep neural network to estimate the Q-function and select actions greedily during training. Double DQN decouples action selection from evaluation to reduce overestimation. Prioritized replay improves sampling to focus on surprising transitions. The Dueling Network separately estimates the state value and state-dependent action advantages to better determine the optimal action. It achieves state-of-the-art performance on Atari games by implicitly splitting credit assignment between choosing now versus later actions.
Kaggle reviewPlanet: Understanding the Amazon from SpaceEduard Tyantov
This document summarizes a Kaggle competition to detect deforestation in the Amazon rainforest using satellite images. It describes:
1. The competition involved classifying over 150,000 image chips into 17 land cover classes to detect deforestation.
2. The baseline model was a ResNet-18 pretrained on ImageNet with fine-tuning, which achieved a score of 90.06%. Several techniques like optimal class thresholds and hyperparameter tuning improved the score to 92.53%.
3. The top models combined RGB satellite images with a near-infrared channel and indexes, training separate branches on JPG and TIF data. The best single model scored 93.071% by ensembling different model
MetaPerturb: Transferable Regularizer for Heterogeneous Tasks and ArchitecturesMLAI2
MetaPerturb is a meta-learned perturbation function that can enhance generalization of neural networks on different tasks and architectures. It proposes a novel meta-learning framework involving jointly training a main model and perturbation module on multiple source tasks to learn a transferable perturbation function. This meta-learned perturbation function can then be transferred to improve performance of a target model on an unseen target task or architecture, outperforming baselines on various datasets and architectures.
Anima Anadkumar, Principal Scientist, Amazon Web Services, Endowed Professor,...MLconf
Large-scale Machine Learning: Deep, Distributed and Multi-Dimensional:
Modern machine learning involves deep neural network architectures which yields state-of-art performance on multiple domains such as computer vision, natural language processing and speech recognition. As the data and models scale, it becomes necessary to have multiple processing units for both training and inference. Apache MXNet is an open-source framework developed for distributed deep learning. I will describe the underlying lightweight hierarchical parameter server architecture that results in high efficiency in distributed settings.
Pushing the current boundaries of deep learning requires using multiple dimensions and modalities. These can be encoded into tensors, which are natural extensions of matrices. We present new deep learning architectures that preserve the multi-dimensional information in data end-to-end. We show that tensor contractions and regression layers are an effective replacement for fully connected layers in deep learning architectures. They result in significant space savings with negligible performance degradation. These functionalities are available in the Tensorly package with MXNet backend interface for large-scale efficient learning.
Bio: Anima Anandkumar is a principal scientist at Amazon Web Services and a Bren professor at Caltech CMS department. Her research interests are in the areas of large-scale machine learning, non-convex optimization and high-dimensional statistics. In particular, she has been spearheading the development and analysis of tensor algorithms. She is the recipient of several awards such as the Alfred. P. Sloan Fellowship, Microsoft Faculty Fellowship, Google research award, ARO and AFOSR Young Investigator Awards, NSF Career Award, Early Career Excellence in Research Award at UCI, Best Thesis Award from the ACM Sigmetrics society, IBM Fran Allen PhD fellowship, and several best paper awards. She has been featured in a number of forums such as the yourstory, Quora ML session, O’Reilly media, and so on. She received her B.Tech in Electrical Engineering from IIT Madras in 2004 and her PhD from Cornell University in 2009. She was a postdoctoral researcher at MIT from 2009 to 2010, an assistant professor at U.C. Irvine between 2010 and 2016, and a visiting researcher at Microsoft Research New England in 2012 and 2014.
This document provides information about a development deep learning architecture event organized by Pantech Solutions and The Institution of Electronics and Telecommunication. The event agenda includes general talks on AI, deep learning libraries, deep learning algorithms like ANN, RNN and CNN, and demonstrations of character recognition and emotion recognition. Details are provided about the organizers Pantech Solutions and IETE, as well as deep learning topics like neural networks, activation functions, common deep learning libraries, algorithms, applications, and the event agenda.
○ 개요
현재 많은 연구자들이 network를 깊고 넓게 설계함으로써 높은 인식률을 갖는 네트워크를 얻고 있다. Network의 크기가 증가하면서 parameter와 computation의 수가 증가하게 되었고, 이러한 문제를 해결하기 위하여 pruning을 기반으로 한 압축 알고리즘들이 제안되어 왔다. 하지만 이러한 방법을 이용하여서는 network architecture자체를 바꿀 수 없기 때문에, 구조에서 오는 한계점들은 해결할 수 없었다.
Network recasting은 구조의 특성으로 인하여 발생하는 한계들을 해결하기 위하여 network architecture 자체를 바꾸는 방법이다. Network recasting을 이용하면 network를 구성하고있는 block들을 다른 형태의 block으로 변환을 할 수 있게 된다. Block-wise recasting 방법을 사용하여 각 block들을 변환할 수 있고, 해당 방법을 연속하여 적용함으로써 전체 network의 구조를 바꿀 수 있다. Sequential recasting 방법을 이용하게 되면 inference accuracy를 더욱 잘 보존할 수 있고, 또한 network architecture에 상관 없이 vanishing gradient problem을 완화 시킬 수 있다. Network recasting을 같은 network architecture에 적용하게 되면 parameter와 computation을 줄이는 효과를 얻을 수 있고, 다른 종류의 network architecture로 변환하게 되면 network를 가속시킬 수 있다. 이러한 경우에는 network architecture 자체를 변경할 수 있기 때문에 구조적 한계보다 더 높은 속도 향상을 얻을 수 있다.
Deep Learning Fast MRI Using Channel Attention in Magnitude DomainJoonhyung Lee
My presentation on how we participated in the fastMRI Challanege in 2019.
Aside from theoretical considerations, it also explains key implementation issues that arise in all deep learning for MRI such as disk I/O and CPU/GPU load balancing.
Used for presentation at ISBI 2020 Oral session.
Accidentally wrote the title as "Deep Learning Sum-of-Squares Images in Accelerated Parallel MRI". Sorry for the mistake!
YOLOv4: optimal speed and accuracy of object detection reviewLEE HOSEONG
YOLOv4 builds upon previous YOLO models and introduces techniques like CSPDarknet53, SPP, PAN, Mosaic data augmentation, and modifications to existing methods to achieve state-of-the-art object detection speed and accuracy while being trainable on a single GPU. Experiments show that combining these techniques through a "bag of freebies" and "bag of specials" approach improves classifier and detector performance over baselines on standard datasets. The paper contributes an efficient object detection model suitable for production use with limited resources.
Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017MLconf
This document discusses deep reinforcement learning and concept network reinforcement learning. It begins with an introduction to reinforcement learning concepts like Markov decision processes and value-based methods. It then describes Concept-Network Reinforcement Learning which decomposes complex tasks into high-level concepts or actions. This allows composing existing solutions to sub-problems without retraining. The document provides examples of using concept networks for lunar lander and robot pick-and-place tasks. It concludes by discussing how concept networks can improve sample efficiency, especially for sparse reward problems.
Restricting the Flow: Information Bottlenecks for Attributiontaeseon ryu
101번째 영상,
펀디멘탈팀 김준호 님의
Restricting the Flow: Information Bottlenecks for Attribution
논문 리뷰 입니다
Explanable ai, xai와 관련된 페이퍼 입니다! 관련되어 관심있으신 분들이 많은 도움이 되시길 바랍니다! attribution map을 이용하여 결과물에 영향을 준 네트워크의 gradient를 직접 추적하여 비주얼 explanation을 추적하는 방식입니다! 펀디멘탈팀 김준호님이 밑바닥부터 자세한 리뷰를 도와주셨습니다!
오늘도 많은 관심과 사랑 감사합니다!
PR-343: Semi-Supervised Semantic Segmentation with Cross Pseudo SupervisionSungchul Kim
This document proposes a new semi-supervised learning method called Cross Pseudo Supervision (CPS) for semantic segmentation. CPS trains two segmentation networks simultaneously where each network generates pseudo labels for the other using its own predictions. It incorporates this cross pseudo supervision loss with a CutMix data augmentation technique. The document evaluates CPS on PASCAL VOC 2012 and Cityscapes datasets with limited labeled data, finding it outperforms other semi-supervised segmentation methods and ablations. Qualitative results show CPS with CutMix generates more accurate segmentations than alternatives.
Human uncertainty makes classification more robust, ICCV 2019 ReviewLEE HOSEONG
1. The document summarizes a research paper that proposes training deep neural networks on soft labels representing human uncertainty in image classification, which improves generalization and robustness compared to training on hard labels.
2. Experiments show that models trained on soft labels constructed from human responses better fit patterns of human uncertainty and improve accuracy, cross-entropy, and a new second-best accuracy measure on various generalization datasets.
3. Alternative soft label methods are also explored, finding that human uncertainty provides a more important contribution than soft labels alone. While robustness to adversarial attacks is improved, defenses are still needed.
201907 AutoML and Neural Architecture SearchDaeJin Kim
Brief introduction of NAS
Review of EfficientNet (Google Brain), RandWire (FAIR) papers
NAS flow slide from KihoSuh's slideshare (https://www.slideshare.net/KihoSuh/neural-architecture-search-with-reinforcement-learning-76883153)
[References]
[1] EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (https://arxiv.org/abs/1905.11946)
[2] Exploring Randomly Wired Neural Networks for Image Recognition (https://arxiv.org/abs/1904.01569)
The document presents research on using neural networks to predict Earth Orientation Parameters (EOP) such as UT1-TAI. Three neural network models were tested:
1) Network 1 varied the number of neurons proportionally with increasing training sample size.
2) Network 2 kept the number of neurons constant while increasing sample size.
3) Network 3 used daily training data with 2 neurons and sample sizes of 4, 10, 20, and 365 days.
The goal was to minimize prediction error (RMSE) for horizons of 5-25 days by adjusting sample size and neurons. Results showed the best balance was needed between these factors, and that short-term prediction was possible within 10 days using
Dueling Network Architectures for Deep Reinforcement LearningYoonho Lee
This document summarizes reinforcement learning algorithms like Deep Q-Network (DQN), Double DQN, prioritized experience replay, and the Dueling Network architecture. DQN uses a deep neural network to estimate the Q-function and select actions greedily during training. Double DQN decouples action selection from evaluation to reduce overestimation. Prioritized replay improves sampling to focus on surprising transitions. The Dueling Network separately estimates the state value and state-dependent action advantages to better determine the optimal action. It achieves state-of-the-art performance on Atari games by implicitly splitting credit assignment between choosing now versus later actions.
Kaggle reviewPlanet: Understanding the Amazon from SpaceEduard Tyantov
This document summarizes a Kaggle competition to detect deforestation in the Amazon rainforest using satellite images. It describes:
1. The competition involved classifying over 150,000 image chips into 17 land cover classes to detect deforestation.
2. The baseline model was a ResNet-18 pretrained on ImageNet with fine-tuning, which achieved a score of 90.06%. Several techniques like optimal class thresholds and hyperparameter tuning improved the score to 92.53%.
3. The top models combined RGB satellite images with a near-infrared channel and indexes, training separate branches on JPG and TIF data. The best single model scored 93.071% by ensembling different model
MetaPerturb: Transferable Regularizer for Heterogeneous Tasks and ArchitecturesMLAI2
MetaPerturb is a meta-learned perturbation function that can enhance generalization of neural networks on different tasks and architectures. It proposes a novel meta-learning framework involving jointly training a main model and perturbation module on multiple source tasks to learn a transferable perturbation function. This meta-learned perturbation function can then be transferred to improve performance of a target model on an unseen target task or architecture, outperforming baselines on various datasets and architectures.
Anima Anadkumar, Principal Scientist, Amazon Web Services, Endowed Professor,...MLconf
Large-scale Machine Learning: Deep, Distributed and Multi-Dimensional:
Modern machine learning involves deep neural network architectures which yields state-of-art performance on multiple domains such as computer vision, natural language processing and speech recognition. As the data and models scale, it becomes necessary to have multiple processing units for both training and inference. Apache MXNet is an open-source framework developed for distributed deep learning. I will describe the underlying lightweight hierarchical parameter server architecture that results in high efficiency in distributed settings.
Pushing the current boundaries of deep learning requires using multiple dimensions and modalities. These can be encoded into tensors, which are natural extensions of matrices. We present new deep learning architectures that preserve the multi-dimensional information in data end-to-end. We show that tensor contractions and regression layers are an effective replacement for fully connected layers in deep learning architectures. They result in significant space savings with negligible performance degradation. These functionalities are available in the Tensorly package with MXNet backend interface for large-scale efficient learning.
Bio: Anima Anandkumar is a principal scientist at Amazon Web Services and a Bren professor at Caltech CMS department. Her research interests are in the areas of large-scale machine learning, non-convex optimization and high-dimensional statistics. In particular, she has been spearheading the development and analysis of tensor algorithms. She is the recipient of several awards such as the Alfred. P. Sloan Fellowship, Microsoft Faculty Fellowship, Google research award, ARO and AFOSR Young Investigator Awards, NSF Career Award, Early Career Excellence in Research Award at UCI, Best Thesis Award from the ACM Sigmetrics society, IBM Fran Allen PhD fellowship, and several best paper awards. She has been featured in a number of forums such as the yourstory, Quora ML session, O’Reilly media, and so on. She received her B.Tech in Electrical Engineering from IIT Madras in 2004 and her PhD from Cornell University in 2009. She was a postdoctoral researcher at MIT from 2009 to 2010, an assistant professor at U.C. Irvine between 2010 and 2016, and a visiting researcher at Microsoft Research New England in 2012 and 2014.
This document provides information about a development deep learning architecture event organized by Pantech Solutions and The Institution of Electronics and Telecommunication. The event agenda includes general talks on AI, deep learning libraries, deep learning algorithms like ANN, RNN and CNN, and demonstrations of character recognition and emotion recognition. Details are provided about the organizers Pantech Solutions and IETE, as well as deep learning topics like neural networks, activation functions, common deep learning libraries, algorithms, applications, and the event agenda.
This document provides legal notices and disclaimers for an informational presentation by Intel. It states that the presentation is for informational purposes only and that Intel makes no warranties. It also notes that Intel technologies' features and benefits depend on system configuration. Finally, it specifies that the sample source code in the presentation is released under the Intel Sample Source Code License Agreement and that Intel and its logo are trademarks.
This document provides an introduction to deep learning. It begins with an overview of artificial intelligence techniques like computer vision, speech processing, and natural language processing that benefit from deep learning. It then reviews the history of deep learning algorithms from perceptrons to modern deep neural networks. The core concepts of deep learning processes, neural network architectures, and training techniques like backpropagation are explained. Popular deep learning frameworks like TensorFlow, Keras, and PyTorch are also introduced. Finally, examples of convolutional neural networks, recurrent neural networks, and generative adversarial networks are briefly described along with tips for training deep neural networks and resources for further learning.
Deep Learning Interview Questions And Answers | AI & Deep Learning Interview ...Simplilearn
- TensorFlow is a popular deep learning library that provides both C++ and Python APIs to make working with deep learning models easier. It supports both CPU and GPU computing and has a faster compilation time than other libraries like Keras and Torch.
- Tensors are multidimensional arrays that represent inputs, outputs, and parameters of deep learning models in TensorFlow. They are the fundamental data structure that flows through graphs in TensorFlow.
- The main programming elements in TensorFlow include constants, variables, placeholders, and sessions. Constants are parameters whose values do not change, variables allow adding trainable parameters, placeholders feed data from outside the graph, and sessions run the graph to evaluate nodes.
Artificial neural network model & hidden layers in multilayer artificial neur...Muhammad Ishaq
Artificial neural networks (ANNs) are computational models inspired by biological neural networks. ANNs can process large amounts of inputs to learn from data in a way similar to the human brain. There are different types of ANN architectures including single layer feedforward networks, multilayer feedforward networks, and recurrent networks. ANNs use supervised, unsupervised, or reinforced learning. The backpropagation algorithm is commonly used for training multilayer networks by propagating errors backwards from the output to adjust weights. Developing an ANN application involves collecting data, separating it into training and testing sets, designing the network architecture, initializing parameters/weights, transforming data, training the network using an algorithm like backpropagation, testing performance on new data, and
This document discusses deep learning initiatives at NECSTLab focused on hardware acceleration of convolutional neural networks using FPGAs. It proposes a framework called CNNECST that provides high-level APIs to design CNNs, integrates with machine learning frameworks for training, and generates customized hardware for FPGA implementation through C++ libraries and Vivado. Experimental results show speedups and energy savings for CNNs like LeNet and MNIST on FPGA boards compared to CPU. Challenges and future work include supporting more layer types and reduced precision computations.
Machine Learning, Deep Learning and Data Analysis IntroductionTe-Yen Liu
The document provides an introduction and overview of machine learning, deep learning, and data analysis. It discusses key concepts like supervised and unsupervised learning. It also summarizes the speaker's experience taking online courses and studying resources to learn machine learning techniques. Examples of commonly used machine learning algorithms and neural network architectures are briefly outlined.
The document provides an introduction to deep learning and how to compute gradients in deep learning models. It discusses machine learning concepts like training models on data to learn patterns, supervised learning tasks like image classification, and optimization techniques like stochastic gradient descent. It then explains how to compute gradients using backpropagation in deep multi-layer neural networks, allowing models to be trained on large datasets. Key steps like the chain rule and backpropagation of errors from the final layer back through the network are outlined.
The document presents a project on sentiment analysis of human emotions, specifically focusing on detecting emotions from babies' facial expressions using deep learning. It involves loading a facial expression dataset, training a convolutional neural network model to classify 7 emotions (anger, disgust, fear, happy, sad, surprise, neutral), and evaluating the model on test data. An emotion detection application is implemented using the trained model to analyze emotions in real-time images from a webcam with around 60-70% accuracy on random images.
This document provides an overview of computer vision techniques including classification and object detection. It discusses popular deep learning models such as AlexNet, VGGNet, and ResNet that advanced the state-of-the-art in image classification. It also covers applications of computer vision in areas like healthcare, self-driving cars, and education. Additionally, the document reviews concepts like the classification pipeline in PyTorch, data augmentation, and performance metrics for classification and object detection like precision, recall, and mAP.
This document provides an overview of deep learning concepts including neural networks, regression and classification, convolutional neural networks, and applications of deep learning such as housing price prediction. It discusses techniques for training neural networks including feature extraction, cost functions, gradient descent, and regularization. The document also reviews deep learning frameworks and notable deep learning models like AlexNet that have achieved success in tasks such as image classification.
Finding the best solution for Image ProcessingTech Triveni
What is beyond using Tensorflow, GPU or TPU to process images seamlessly? Do we have a silver bullet for image processing? Over the years, image processing has picked up a different level of attraction. Everyone can think about its ease of usability because it has become a reality now. We have started seeing how Residual Neural Network architecture is being used for different cases and not only that, how Residual Neural network is being tweaked to solve different problems. Along with tweaking the ResNet, preprocessing is also being improved to support different architecture for this matter.
Everyone has almost become cyborg already with mobile phones in our hands and apparently until human beings bring the AI/ML to the phones completely they are not taking any rest. We are going to see the development of different architecture and algorithms around running AI/ML on low configuration devices.
In this session, we are going to talk about different research papers submitted for these matters and some implementations for the same as well.
This document provides an overview of deep learning including why it is used, common applications, strengths and challenges, common algorithms, and techniques for developing deep learning models. In 3 sentences: Deep learning methods like neural networks can learn complex patterns in large, unlabeled datasets and are better than traditional machine learning for tasks like image recognition. Popular deep learning algorithms include convolutional neural networks for image data and recurrent neural networks for sequential data. Effective deep learning requires techniques like regularization, dropout, data augmentation, and hyperparameter optimization to prevent overfitting on training data.
This document provides an overview of deep learning including:
1. Why deep learning performs better than traditional machine learning for tasks like image and speech recognition.
2. Common deep learning applications such as image recognition, speech recognition, and healthcare.
3. Challenges of deep learning like the need for large datasets and lack of interpretability.
In this talk, after a brief overview of AI concepts in particular Machine Learning (ML) techniques, some of the well-known computer design concepts for high performance and power efficiency are presented. Subsequently, those techniques that have had a promising impact for computing ML algorithms are discussed. Deep learning has emerged as a game changer for many applications in various fields of engineering and medical sciences. Although the primary computation function is matrix vector multiplication, many competing efficient implementations of this primary function have been proposed and put into practice. This talk will review and compare some of those techniques that are used for ML computer design.
Deep Learning for Computer Vision - PyconDE 2017Alex Conway
This document discusses deep learning for computer vision tasks. It begins with an overview of image classification using convolutional neural networks and how they have achieved superhuman performance on ImageNet. It then covers the key layers and concepts in CNNs, including convolutions, max pooling, and transferring learning to new problems. Finally, it discusses more advanced computer vision tasks that CNNs have been applied to, such as semantic segmentation, style transfer, visual question answering, and combining images with other data sources.
Activation functions and Training Algorithms for Deep Neural networkGayatri Khanvilkar
Training of Deep neural network is difficult task. Deep neural network train with the help of training algorithms and activation function This is an overview of Activation Function and Training Algorithms used for Deep Neural Network. It underlines a brief comparative study of activation function and training algorithms.
Similar to Image classification with neural networks (20)
LAND USE LAND COVER AND NDVI OF MIRZAPUR DISTRICT, UPRAHUL
This Dissertation explores the particular circumstances of Mirzapur, a region located in the
core of India. Mirzapur, with its varied terrains and abundant biodiversity, offers an optimal
environment for investigating the changes in vegetation cover dynamics. Our study utilizes
advanced technologies such as GIS (Geographic Information Systems) and Remote sensing to
analyze the transformations that have taken place over the course of a decade.
The complex relationship between human activities and the environment has been the focus
of extensive research and worry. As the global community grapples with swift urbanization,
population expansion, and economic progress, the effects on natural ecosystems are becoming
more evident. A crucial element of this impact is the alteration of vegetation cover, which plays a
significant role in maintaining the ecological equilibrium of our planet.Land serves as the foundation for all human activities and provides the necessary materials for
these activities. As the most crucial natural resource, its utilization by humans results in different
'Land uses,' which are determined by both human activities and the physical characteristics of the
land.
The utilization of land is impacted by human needs and environmental factors. In countries
like India, rapid population growth and the emphasis on extensive resource exploitation can lead
to significant land degradation, adversely affecting the region's land cover.
Therefore, human intervention has significantly influenced land use patterns over many
centuries, evolving its structure over time and space. In the present era, these changes have
accelerated due to factors such as agriculture and urbanization. Information regarding land use and
cover is essential for various planning and management tasks related to the Earth's surface,
providing crucial environmental data for scientific, resource management, policy purposes, and
diverse human activities.
Accurate understanding of land use and cover is imperative for the development planning
of any area. Consequently, a wide range of professionals, including earth system scientists, land
and water managers, and urban planners, are interested in obtaining data on land use and cover
changes, conversion trends, and other related patterns. The spatial dimensions of land use and
cover support policymakers and scientists in making well-informed decisions, as alterations in
these patterns indicate shifts in economic and social conditions. Monitoring such changes with the
help of Advanced technologies like Remote Sensing and Geographic Information Systems is
crucial for coordinated efforts across different administrative levels. Advanced technologies like
Remote Sensing and Geographic Information Systems
9
Changes in vegetation cover refer to variations in the distribution, composition, and overall
structure of plant communities across different temporal and spatial scales. These changes can
occur natural.
The simplified electron and muon model, Oscillating Spacetime: The Foundation...RitikBhardwaj56
Discover the Simplified Electron and Muon Model: A New Wave-Based Approach to Understanding Particles delves into a groundbreaking theory that presents electrons and muons as rotating soliton waves within oscillating spacetime. Geared towards students, researchers, and science buffs, this book breaks down complex ideas into simple explanations. It covers topics such as electron waves, temporal dynamics, and the implications of this model on particle physics. With clear illustrations and easy-to-follow explanations, readers will gain a new outlook on the universe's fundamental nature.
How to Manage Your Lost Opportunities in Odoo 17 CRMCeline George
Odoo 17 CRM allows us to track why we lose sales opportunities with "Lost Reasons." This helps analyze our sales process and identify areas for improvement. Here's how to configure lost reasons in Odoo 17 CRM
This presentation was provided by Steph Pollock of The American Psychological Association’s Journals Program, and Damita Snow, of The American Society of Civil Engineers (ASCE), for the initial session of NISO's 2024 Training Series "DEIA in the Scholarly Landscape." Session One: 'Setting Expectations: a DEIA Primer,' was held June 6, 2024.
How to Build a Module in Odoo 17 Using the Scaffold MethodCeline George
Odoo provides an option for creating a module by using a single line command. By using this command the user can make a whole structure of a module. It is very easy for a beginner to make a module. There is no need to make each file manually. This slide will show how to create a module using the scaffold method.
বাংলাদেশের অর্থনৈতিক সমীক্ষা ২০২৪ [Bangladesh Economic Review 2024 Bangla.pdf] কম্পিউটার , ট্যাব ও স্মার্ট ফোন ভার্সন সহ সম্পূর্ণ বাংলা ই-বুক বা pdf বই " সুচিপত্র ...বুকমার্ক মেনু 🔖 ও হাইপার লিংক মেনু 📝👆 যুক্ত ..
আমাদের সবার জন্য খুব খুব গুরুত্বপূর্ণ একটি বই ..বিসিএস, ব্যাংক, ইউনিভার্সিটি ভর্তি ও যে কোন প্রতিযোগিতা মূলক পরীক্ষার জন্য এর খুব ইম্পরট্যান্ট একটি বিষয় ...তাছাড়া বাংলাদেশের সাম্প্রতিক যে কোন ডাটা বা তথ্য এই বইতে পাবেন ...
তাই একজন নাগরিক হিসাবে এই তথ্য গুলো আপনার জানা প্রয়োজন ...।
বিসিএস ও ব্যাংক এর লিখিত পরীক্ষা ...+এছাড়া মাধ্যমিক ও উচ্চমাধ্যমিকের স্টুডেন্টদের জন্য অনেক কাজে আসবে ...
हिंदी वर्णमाला पीपीटी, hindi alphabet PPT presentation, hindi varnamala PPT, Hindi Varnamala pdf, हिंदी स्वर, हिंदी व्यंजन, sikhiye hindi varnmala, dr. mulla adam ali, hindi language and literature, hindi alphabet with drawing, hindi alphabet pdf, hindi varnamala for childrens, hindi language, hindi varnamala practice for kids, https://www.drmullaadamali.com
How to Make a Field Mandatory in Odoo 17Celine George
In Odoo, making a field required can be done through both Python code and XML views. When you set the required attribute to True in Python code, it makes the field required across all views where it's used. Conversely, when you set the required attribute in XML views, it makes the field required only in the context of that particular view.
Main Java[All of the Base Concepts}.docxadhitya5119
This is part 1 of my Java Learning Journey. This Contains Custom methods, classes, constructors, packages, multithreading , try- catch block, finally block and more.
1. Amirkabir University of Technology
Department of Computer Engineering
and Information Technology
Image Classification with
Deep Convolutional
Neural Networks
Sepehr Rasouli
2. Introduction > Methods > Results > Conclusion2
Outline
• Introduction to Image Classification
& Deep Networks
• Proposed Method
• Main Idea
• Data Set
• Architecture
• Techniques
• Comparison & Results
• Conclusion
4. Introduction > Methods > Results > Conclusion4
Why Deep Learning?
•“Shallow” vs. “deep” architectures
Learn a feature hierarchy all the way from pixels to classifier
Hand Designed
Feature
Extraction
Trainable
Classifier
Layer 1 Layer N
Simpler
classifier
5. Introduction > Methods > Results > Conclusion5
Our Method
• Deep Convolutional Neural Network
• 5 convolutional and 3 fully connected layers
• 650,000 neurons, 60 million parameters
• Techniques used for boosting up performance
• ReLU nonlinearity
• Training on Multiple GPUs
• Overlapping max pooling
• Data Augmentation
• Dropout
6. Introduction > Methods > Results > Conclusion6
Overall Architecture
• Trained with stochastic gradient descent on two NVIDIA GPUs for about a
week (5~6 days)
• 650,000 neurons, 60 million parameters, 630 million connections
• The last layer contains 1,000 neurons which produces a distribution over the
1,000 class labels.
7. Introduction > Methods > Results > Conclusion7
Dataset
• ImageNet
§ Over 15 million high-quality labeled images
§ About 22,000 categories
§ Collected from the web, labeled by humans on Amazon's Mechanical
Turk
§ Variable-resolution images
• ILSVRC Competition
§ ImageNet Large-Scale Pascal Visual Object Challenge
§ Annual competition of image classification at large scale
§ Subset of ImageNet
§ 1,000 categories with about 1,000 images each
§ 1.2M images in 1K categories
§ Classification: make 5 guesses about the image label
8. Introduction > Methods > Results > Conclusion8
Rectified Linear Units
𝑥 = 𝑤$ 𝑓 𝑍$ + 𝑤( 𝑓 𝑍(
+𝑤) 𝑓 𝑍)
x is called the total input
to the neuron, and f(x)
is its output
Very bad
(slow to train )
Very good
(quick to train)
f(x) = max(0,x)f(x) = tanh(x)
9. Introduction > Methods > Results > Conclusion9
Rectified Linear Units
• Biological plausibility: One-sided, compared
to the antisymmetry of tanh.
• Sparse activation: For example, in a randomly
initialized network, only about 50% of hidden
units are activated (having a non-zero output).
• Efficient gradient propagation: No vanishing
gradient problem or exploding effect.
• Efficient computation: Only comparison,
addition and multiplication
10. Introduction > Methods > Results > Conclusion10
Training on Multiple GPUs
• Spread across two GPUs
• GTX 580 GPU with 3GB memory
• Particularly well-suited to cross-GPU
parallelization
• Very efficient implementation of CNN on
GPUs
11. Model Top-1 Top-5
Sparse coding [3] 47.1% 28.2%
SIFT + FVs [4] 45.7% 25.7%
CNN 37.5 17.0%
Introduction > Methods > Results > Conclusion11
Results & Comparison
•ILSVRC-2010 test set
ILSVRC-2010 winner
Previous best
published result
Our Method
Comparison of results on ILSRVCs 2010
test set. In italics best results achieved
by others.
12. Introduction > Methods > Results > Conclusion12
Conclusion
• Large, deep convolutional neural networks for large
scale image classification was proposed
• 5 convolutional layers, 3 fully-connected layers
• 650,000 neurons, 60 million parameters
• Several techniques for boosting up performance
• The proposed method won the ILSVRC-2012
• Achieved a winning top-5 error rate of 15.3%,
compared to 26.2% achieved by the second-best entry
14. Introduction > Methods > Results > Conclusion14
References
[1] http://cs.nyu.edu/~fergus/tutorials/
deep_learning_cvpr12/fergus_dl_tutorial_final.pptx
[2] reference : http://web.engr.illinois.edu/
~slazebni/spring14/lec24_cnn.pdf
[3] A. Berg, J. Deng, and L. Fei-Fei. Large scale
visual recognition challenge 2010.
www.imagenet.org/challenges. 2010. [4]
S. Tara, Brian Kingsbury, A.-r. Mohamed and
B. Ramabhadran, "Learning Filter Banks within a Deep
[4] J.Sánchezand F.Perronnin.High-dimensional
signature compression for large-scale image classification.
In Computer Vision and Pattern Recognition(CVPR),
2011IEEEConferenceon,pages1665–1672.IEEE, 2011.
15. Introduction > Methods > Results > Conclusion15
Thank you for your attention
Any Questions?
18. Introduction > Methods > Results > Conclusion18
Pooling
• Spatial Pooling
• Non-overlapping / overlapping regions
• Sum or max
Max
Sum
19. Introduction > Methods > Results > Conclusion19
Dropout
• Independently set each hidden unit activity to zero with 0.5
probability
• Used in the two globally-connected hidden layers at the net's
output