�ݺ�ߣshows by User: AnimaAnandkumar

�ݺ�ߣshows by User: AnimaAnandkumar / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: AnimaAnandkumar / Tue, 29 May 2018 21:23:54 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: AnimaAnandkumar Role of Tensors in Machine Learning /slideshow/role-of-tensors-in-machine-learning/99428003 scaled-ml-2018-180529212354
Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. Tensors play a significant role in machine learning through (1) tensor contractions, (2) tensor sketches, and (3) tensor decompositions. Tensor contractions are extensions of matrix products to higher dimensions. Tensor sketches efficiently compress tensors while preserving information. Tensor decompositions compute low rank components that constitute a tensor.]]>
Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. Tensors play a significant role in machine learning through (1) tensor contractions, (2) tensor sketches, and (3) tensor decompositions. Tensor contractions are extensions of matrix products to higher dimensions. Tensor sketches efficiently compress tensors while preserving information. Tensor decompositions compute low rank components that constitute a tensor.]]> Tue, 29 May 2018 21:23:54 GMT /slideshow/role-of-tensors-in-machine-learning/99428003 AnimaAnandkumar@slideshare.net(AnimaAnandkumar) Role of Tensors in Machine Learning AnimaAnandkumar Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. Tensors play a significant role in machine learning through (1) tensor contractions, (2) tensor sketches, and (3) tensor decompositions. Tensor contractions are extensions of matrix products to higher dimensions. Tensor sketches efficiently compress tensors while preserving information. Tensor decompositions compute low rank components that constitute a tensor. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/scaled-ml-2018-180529212354-thumbnail.jpg?width=120&height=120&fit=bounds" /><br> Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. Tensors play a significant role in machine learning through (1) tensor contractions, (2) tensor sketches, and (3) tensor decompositions. Tensor contractions are extensions of matrix products to higher dimensions. Tensor sketches efficiently compress tensors while preserving information. Tensor decompositions compute low rank components that constitute a tensor.

Role of Tensors in Machine Learning from Anima Anandkumar

]]> 7112 4 https://cdn.slidesharecdn.com/ss_thumbnails/scaled-ml-2018-180529212354-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Trinity of AI: data, algorithms and cloud /slideshow/trinity-of-ai-data-algorithms-and-cloud-97455108/97455108 trinity-180518052457
AI at scale requires a perfect storm of data, algorithms and cloud infrastructure. Modern deep learning requires large amounts of training data. We develop methods that improve data collection, aggregation and augmentation. This involves active learning, partial feedback, crowdsourcing, and generative models. We analyze large-scale machine learning methods for distributed training. We show that gradient quantization can yield best of both the worlds: accuracy and communication efficiency. We extend matrix methods to higher dimensions using tensor algebraic techniques and show superior performance. Finally, at AWS, we are developing robust software frameworks and AI cloud services at all levels of the stack.]]>
AI at scale requires a perfect storm of data, algorithms and cloud infrastructure. Modern deep learning requires large amounts of training data. We develop methods that improve data collection, aggregation and augmentation. This involves active learning, partial feedback, crowdsourcing, and generative models. We analyze large-scale machine learning methods for distributed training. We show that gradient quantization can yield best of both the worlds: accuracy and communication efficiency. We extend matrix methods to higher dimensions using tensor algebraic techniques and show superior performance. Finally, at AWS, we are developing robust software frameworks and AI cloud services at all levels of the stack.]]> Fri, 18 May 2018 05:24:57 GMT /slideshow/trinity-of-ai-data-algorithms-and-cloud-97455108/97455108 AnimaAnandkumar@slideshare.net(AnimaAnandkumar) Trinity of AI: data, algorithms and cloud AnimaAnandkumar AI at scale requires a perfect storm of data, algorithms and cloud infrastructure. Modern deep learning requires large amounts of training data. We develop methods that improve data collection, aggregation and augmentation. This involves active learning, partial feedback, crowdsourcing, and generative models. We analyze large-scale machine learning methods for distributed training. We show that gradient quantization can yield best of both the worlds: accuracy and communication efficiency. We extend matrix methods to higher dimensions using tensor algebraic techniques and show superior performance. Finally, at AWS, we are developing robust software frameworks and AI cloud services at all levels of the stack. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/trinity-180518052457-thumbnail.jpg?width=120&height=120&fit=bounds" /><br> AI at scale requires a perfect storm of data, algorithms and cloud infrastructure. Modern deep learning requires large amounts of training data. We develop methods that improve data collection, aggregation and augmentation. This involves active learning, partial feedback, crowdsourcing, and generative models. We analyze large-scale machine learning methods for distributed training. We show that gradient quantization can yield best of both the worlds: accuracy and communication efficiency. We extend matrix methods to higher dimensions using tensor algebraic techniques and show superior performance. Finally, at AWS, we are developing robust software frameworks and AI cloud services at all levels of the stack.

Trinity of AI: data, algorithms and cloud from Anima Anandkumar

]]> 2375 6 https://cdn.slidesharecdn.com/ss_thumbnails/trinity-180518052457-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Tensors for topic modeling and deep learning on AWS Sagemaker /slideshow/tensors-for-topic-modeling-and-deep-learning-on-aws-sagemaker/83018553 venpalpwed1515mcl337r1llb-171130011620
Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. This session will present recently developed tensor algorithms for topic modeling and deep learning with vastly improved performance over existing methods. Topic models enable automated categorization of large document corpora, without requiring labeled data for training. They go beyond simple clustering since they allow for documents to have multiple topics. Tensor methods provide a fast and a guaranteed method for training these models. They incorporate co-occurrence statistics of triplets of words in documents. We are releasing a fast and a robust implementation that vastly outperform existing solutions while providing significantly faster training times and better topic quality. Moreover, training and inference are decoupled in our algorithm, so the user can select the relevant part based on their requirements. We will present benchmarks across multiple datasets of different sizes and AWS instance types, and provide notebook examples. ]]>
Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. This session will present recently developed tensor algorithms for topic modeling and deep learning with vastly improved performance over existing methods. Topic models enable automated categorization of large document corpora, without requiring labeled data for training. They go beyond simple clustering since they allow for documents to have multiple topics. Tensor methods provide a fast and a guaranteed method for training these models. They incorporate co-occurrence statistics of triplets of words in documents. We are releasing a fast and a robust implementation that vastly outperform existing solutions while providing significantly faster training times and better topic quality. Moreover, training and inference are decoupled in our algorithm, so the user can select the relevant part based on their requirements. We will present benchmarks across multiple datasets of different sizes and AWS instance types, and provide notebook examples. ]]> Thu, 30 Nov 2017 01:16:20 GMT /slideshow/tensors-for-topic-modeling-and-deep-learning-on-aws-sagemaker/83018553 AnimaAnandkumar@slideshare.net(AnimaAnandkumar) Tensors for topic modeling and deep learning on AWS Sagemaker AnimaAnandkumar Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. This session will present recently developed tensor algorithms for topic modeling and deep learning with vastly improved performance over existing methods. Topic models enable automated categorization of large document corpora, without requiring labeled data for training. They go beyond simple clustering since they allow for documents to have multiple topics. Tensor methods provide a fast and a guaranteed method for training these models. They incorporate co-occurrence statistics of triplets of words in documents. We are releasing a fast and a robust implementation that vastly outperform existing solutions while providing significantly faster training times and better topic quality. Moreover, training and inference are decoupled in our algorithm, so the user can select the relevant part based on their requirements. We will present benchmarks across multiple datasets of different sizes and AWS instance types, and provide notebook examples. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/venpalpwed1515mcl337r1llb-171130011620-thumbnail.jpg?width=120&height=120&fit=bounds" /><br> Tensors are higher order extensions of matrices that can incorporate multiple modalities and encode higher order relationships in data. This session will present recently developed tensor algorithms for topic modeling and deep learning with vastly improved performance over existing methods. Topic models enable automated categorization of large document corpora, without requiring labeled data for training. They go beyond simple clustering since they allow for documents to have multiple topics. Tensor methods provide a fast and a guaranteed method for training these models. They incorporate co-occurrence statistics of triplets of words in documents. We are releasing a fast and a robust implementation that vastly outperform existing solutions while providing significantly faster training times and better topic quality. Moreover, training and inference are decoupled in our algorithm, so the user can select the relevant part based on their requirements. We will present benchmarks across multiple datasets of different sizes and AWS instance types, and provide notebook examples.

Tensors for topic modeling and deep learning on AWS Sagemaker from Anima Anandkumar

]]> 2794 5 https://cdn.slidesharecdn.com/ss_thumbnails/venpalpwed1515mcl337r1llb-171130011620-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://public.slidesharecdn.com/v2/images/profile-picture.png https://cdn.slidesharecdn.com/ss_thumbnails/scaled-ml-2018-180529212354-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/role-of-tensors-in-machine-learning/99428003 Role of Tensors in Mac... https://cdn.slidesharecdn.com/ss_thumbnails/trinity-180518052457-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/trinity-of-ai-data-algorithms-and-cloud-97455108/97455108 Trinity of AI: data, a... https://cdn.slidesharecdn.com/ss_thumbnails/venpalpwed1515mcl337r1llb-171130011620-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/tensors-for-topic-modeling-and-deep-learning-on-aws-sagemaker/83018553 Tensors for topic mode...