A stroke occurs when the blood supply to the brain is obstructed. Sometimes a stroke can cause long-term disability. And the path to stroke recovery is long and arduous.

The stroke rehab effect is poor, and the patient’s willingness to train is likewise low. As stroke affects different brain systems, patients who have undergone stroke rehabilitation show a “heterogeneity in the outcome,” a medical term that suggests stroke recovery treatment can be very different between individual stroke victims.

Maximizing individual stroke recovery treatment requires finding the optimal neuro-rehabilitative strategy.

Dr. Philipp J. Koch, the study’s first author, said, “If we want to address these challenges in everyday clinical practice, we have first to enhance our ability to predict the individual courses of recovery.”

In a new study, an international team of scientists led by EPFL has developed a system that combines information from the brain’s connectome and machine learning to assess and predict the outcome of stroke victims.

For neuroscientists, connectomes act as an indispensable tool, especially when interpreting structural or dynamic brain data and associating them with functions, functional deficits, or recovery processes.

For the study, scientists analyzed connectomes from 92 patients two weeks after the stroke. They tracked changes in connectome up to three months later while assessing motor impairment with a standardized scale. This allowed them to monitor connection changes in the individual brains of the patients while they underwent recovery.

The scientists input the connectome information into a “support-vector machine,” or SVM, a type of machine-learning model that uses examples to map an input onto an output.

The SVM was trained to distinguish between patients with natural recovery from those without their whole-brain structural connectomes. The SVMs then characterized every patient’s basic brain network pattern, focusing on the individuals who were seriously debilitated to make predictions about their recovery potential, with the accuracy of each prediction cross-validated internally and externally with independent datasets.

A result is a cutting-edge tool of personalized medicine: a machine-learning system that can identify neuronal network patterns to make high-accuracy predictions on the outcome of recovery for stroke patients.

Professor Friedhelm Hummel, a neuroscientist and Director of the Defitech Chair for Clinical Neuroengineering at EPFL’s School of Life Sciences, said, “This tool can support the prediction of individual courses of recovery early on and will have an important impact on clinical management, translational research, and treatment choice.”

The post Improving stroke recovery prediction using machine learning appeared first on Artificial Intelligence.

• AI and Machine Learning are on track to generate between $1.4 Trillion to $2.6 Trillion in value by solving Marketing and Sales problems over the next three years, according to the McKinsey Global Institute. 
• Marketers’ use of AI soared between 2018 and 2020, jumping from 29% in 2018 to 84% in 2020, according to Salesforce Research’s most recent State of Marketing Study. 
• AI, Machine Learning, marketing & advertising technologies, voice/chat/digital assistants and mobile tech & apps are the five technologies that will have the greatest impact on the future of marketing, according to Drift’s 2020 Marketing Leadership Benchmark Report.

Chief Marketing Officers (CMOs) and the marketing teams they lead are expected to excel at creating customer trust, a brand that exudes empathy and data-driven strategies that deliver results. Personalizing channel experiences at scale works when CMOs strike the perfect balance between their jobs’ emotional and logical, data-driven parts. That’s what makes being a CMO today so challenging. They’ve got to have the compassion of a Captain Kirk and the cold, hard logic of a Dr. Spock and know when to use each skill set. CMOs and their teams struggle to keep the emotional and logical parts of their jobs in balance. Machine learning apps and platforms are helping to acheive goals in both side of their roles.

Asked how her team keeps them in balance, the CMO of an enterprise software company told me she always leads with empathy, safety and security for customers and results follow. “Throughout the pandemic, our message to our customers is that their health and safety come first and we’ll provide additional services at no charge if they need it.” True to her word, the company offered their latest cybersecurity release update to all customers free in 2020.  AI and machine learning tools help her and her team test, learn and excel iteratively to create an empathic brand that delivers results.

The following are ten ways AI and machine learning are improving marketing in 2021:

1.    70% of high-performance marketing teams claim they have a fully defined AI strategy versus 35% of their under-performing peer marketing team counterparts. CMOs who lead high-performance marketing teams place a high value on continually learning and embracing a growth mindset, as evidenced by 56% of them planning to use AI and machine learning over the next year. Choosing to put in the work needed to develop new AI and machine learning skills pays off with improved social marketing performance and greater precision with marketing analytics. Source: State of Marketing, Sixth Edition. Salesforce Research, 2020.

2.    36% of marketers predict AI will have a significant impact on marketing performance this year. 32% of marketers and agency professionals were using AI to create ads, including digital banners, social media posts and digital out-of-home ads, according to a recent study by Advertiser Perceptions. Source: Which Emerging Tech Do Marketers Think Will Most Impact Strategy This Year?, Marketing Charts, January 5, 2021.

3.    High-performing marketing teams are averaging seven different uses of AI and machine learning today and just over half (52%) plan on increasing their adoption this year. High-performing marketing teams and the CMOs lead them to invest in AI and machine learning to improve customer segmentation. They’re also focused on personalizing individual channel experiences. The following graphic underscores how quickly high-performing marketing teams learn then adopt advanced AI and machine learning techniques to their competitive advantage. Source: State of Marketing, Sixth Edition. Salesforce Research, 2020.

4.    Marketers use AI-based demand sensing to better predict unique buying patterns across geographic regions and alleviate stock-outs and back-orders. Combining all available data sources, including customer sentiment analysis using supervised machine learning algorithms, it’s possible to improve demand sensing and demand forecast accuracy. ML algorithms can correlate location-specific sentiment for a given product or brand and a given product’s regional availability. Having this insight alone can save the retail industry up to $50B a year in obsoleted inventory.  Source: AI can help retailers understand the consumer, Phys.org. January 14, 2019.

5.    Disney is applying AI modeling techniques, including machine learning algorithms, to fine-tune and optimize its media mix model. Disney’s approach to gaining new insights into its media mix model is to aggregate data from across the organization including partners, prepare the model data and then transform it for use in a model. Next, a variety of models are used to achieve budget and media mix optimization. Then compare scenarios. The result is a series of insights that are presented to senior management. The following dashboard shows the structure of how they analyze AI-based data internally. The data shown is, for example only; this does not reflect Disney’s actual operations.   Source: How Disney uses Tableau to visualize its media mix model (https://www.tableau.com/best-marketing-dashboards)

6.    41% of marketers say that AI and machine learning make their greatest contributions to accelerating revenue growth and improving performance. Marketers say that getting more actionable insights from marketing data (40%) and creating personalized consumer experiences at scale (38%) round out the top three uses today. The study also found that most marketers, 77%, have less than a quarter of all marketing tasks intelligently automated and 18% say they haven’t intelligently automated any tasks at all. Marketers need to look to AI and machine learning to automated remote, routine tasks to free up more time to create new campaigns. Source: Drift and Marketing Artificial Intelligence Institute, 2021 State of Marketing AI Report.

7.    Starbucks set the ambitious goal of being the world’s most personalized brand by relying on predictive analytics and machine learning to create a real-time personalization experience. The global coffee chain faced several challenges starting with how difficult it was to target individual customers with their existing IT infrastructure. They were also heavily reliant on manual operations across their thousands of stores, which made personalization at scale a formidable challenge to overcome. Starbucks created a real-time personalization engine that integrated with customers’ account information, the mobile app, customer preferences, 3^rd party data and contextual data. They achieved a 150% increase in user interaction using predictive analytics and AI, a 3X improvement in per-customer net incremental revenues. The following is a diagram of how DigitalBCG (Boston Consulting Group) was able to assist them. Source: Becoming The World’s Most Personalized Brand, DigitalBCG. 

8.    Getting personalization-at-scale right starts with a unified Customer Data Platform (CDP) that can use machine learning algorithms to discover new customer data patterns and “learn” over time.  For high-achieving marketing organizations, achieving personalization-at-scale is their highest and most urgent priority based on Salesforce Research’s most recent State of Marketing survey. And McKinsey predicts personalization-at-scale can create $1.7 trillion to $3 trillion in new value. For marketers to capture a part of this value, changes to the mar-tech stack (shown below) must be supported by clear accountability and ownership of channel and customer results. Combining a modified mar-tech stack with clear accountability delivers results.   Source: McKinsey & Company, A technology blueprint for personalization at scale. May 20, 2019. By Sean Flavin and Jason Heller.

9.    Campaign management, mobile app technology and testing/optimization are the leading three plans for a B2C company’s personalization technologies. Just 19% of enterprises have adopted AI and machine learning for B2C personalization today. The Forrester Study commissioned by IBM also found that 55% of enterprises believe the technology limitations inhibit their ability to execute personalization strategies. Source: A Forrester Consulting Thought Leadership Paper, Commissioned by IBM, Personalization Demystified: Enchant Your Customers By Going From Good To Great, February 2020.

10. Successful AI-driven personalization strategies deliver results beyond marketing, delivering strong results enterprise-wide, including lifting sales revenue, Net Promoter Scores and customer retention rates. When personalization-at-scale is done right, enterprises achieve a net 5.63% increase in sales revenue, 10.26% increase in order frequency, uplifts in average order value and an impressive 13.25% improvement in cross-sell/up-sell opportunities. The benefits transcend marketing alone and drive higher customer satisfaction metrics as well.   Source: A Forrester Consulting Thought Leadership Paper, Commissioned by IBM, Personalization Demystified: Enchant Your Customers By Going From Good To Great, February 2020.

CMOs and their teams rely on AI and machine learning to iteratively test and improve every aspect of their marketing campaigns and strategies. Striking the perfect balance between empathy and data-driven results takes a new level of data quality which isn’t possible to achieve using Microsoft Excel or personal productivity tools today. The most popular use of AI and machine learning in organizations is delivering personalization at scale across all digital channels. There’s also increasing adoption of predictive analytics based on machine learning to fine-tune propensity models to improve up-sell and cross-sell results. 

The post 10 Ways AI And Machine Learning Are Improving Marketing In 2021 appeared first on Artificial Intelligence.

• AI and Machine Learning are on track to generate between $1.4 Trillion to $2.6 Trillion in value by solving Marketing and Sales problems over the next three years, according to the McKinsey Global Institute. 
• Marketers’ use of AI soared between 2018 and 2020, jumping from 29% in 2018 to 84% in 2020, according to Salesforce Research’s most recent State of Marketing Study. 
• AI, Machine Learning, marketing & advertising technologies, voice/chat/digital assistants and mobile tech & apps are the five technologies that will have the greatest impact on the future of marketing, according to Drift’s 2020 Marketing Leadership Benchmark Report.

Chief Marketing Officers (CMOs) and the marketing teams they lead are expected to excel at creating customer trust, a brand that exudes empathy and data-driven strategies that deliver results. Personalizing channel experiences at scale works when CMOs strike the perfect balance between their jobs’ emotional and logical, data-driven parts. That’s what makes being a CMO today so challenging. They’ve got to have the compassion of a Captain Kirk and the cold, hard logic of a Dr. Spock and know when to use each skill set. CMOs and their teams struggle to keep the emotional and logical parts of their jobs in balance.

Asked how her team keeps them in balance, the CMO of an enterprise software company told me she always leads with empathy, safety and security for customers and results follow. “Throughout the pandemic, our message to our customers is that their health and safety come first and we’ll provide additional services at no charge if they need it.” True to her word, the company offered their latest cybersecurity release update to all customers free in 2020.  AI and machine learning tools help her and her team test, learn and excel iteratively to create an empathic brand that delivers results.

The following are ten ways AI and machine learning are improving marketing in 2021:

1.    70% of high-performance marketing teams claim they have a fully defined AI strategy versus 35% of their under-performing peer marketing team counterparts. CMOs who lead high-performance marketing teams place a high value on continually learning and embracing a growth mindset, as evidenced by 56% of them planning to use AI and machine learning over the next year. Choosing to put in the work needed to develop new AI and machine learning skills pays off with improved social marketing performance and greater precision with marketing analytics. Source: State of Marketing, Sixth Edition. Salesforce Research, 2020.

What Are The Fastest Growing Cybersecurity Skills In 2021?Top 20 Predictions Of How AI Is Going To Improve Cybersecurity In 2021The Top 20 Cybersecurity Startups To Watch In 2021 Based On Crunchbase

2.    36% of marketers predict AI will have a significant impact on marketing performance this year. 32% of marketers and agency professionals were using AI to create ads, including digital banners, social media posts and digital out-of-home ads, according to a recent study by Advertiser Perceptions. Source: Which Emerging Tech Do Marketers Think Will Most Impact Strategy This Year?, Marketing Charts, January 5, 2021.

3.    High-performing marketing teams are averaging seven different uses of AI and machine learning today and just over half (52%) plan on increasing their adoption this year. High-performing marketing teams and the CMOs lead them to invest in AI and machine learning to improve customer segmentation. They’re also focused on personalizing individual channel experiences. The following graphic underscores how quickly high-performing marketing teams learn then adopt advanced AI and machine learning techniques to their competitive advantage. Source: State of Marketing, Sixth Edition. Salesforce Research, 2020.

4.    Marketers use AI-based demand sensing to better predict unique buying patterns across geographic regions and alleviate stock-outs and back-orders. Combining all available data sources, including customer sentiment analysis using supervised machine learning algorithms, it’s possible to improve demand sensing and demand forecast accuracy. ML algorithms can correlate location-specific sentiment for a given product or brand and a given product’s regional availability. Having this insight alone can save the retail industry up to $50B a year in obsoleted inventory.  Source: AI can help retailers understand the consumer, Phys.org. January 14, 2019.

5.    Disney is applying AI modeling techniques, including machine learning algorithms, to fine-tune and optimize its media mix model. Disney’s approach to gaining new insights into its media mix model is to aggregate data from across the organization including partners, prepare the model data and then transform it for use in a model. Next, a variety of models are used to achieve budget and media mix optimization. Then compare scenarios. The result is a series of insights that are presented to senior management. The following dashboard shows the structure of how they analyze AI-based data internally. The data shown is, for example only; this does not reflect Disney’s actual operations.   Source: How Disney uses Tableau to visualize its media mix model (https://www.tableau.com/best-marketing-dashboards)

6.    41% of marketers say that AI and machine learning make their greatest contributions to accelerating revenue growth and improving performance. Marketers say that getting more actionable insights from marketing data (40%) and creating personalized consumer experiences at scale (38%) round out the top three uses today. The study also found that most marketers, 77%, have less than a quarter of all marketing tasks intelligently automated and 18% say they haven’t intelligently automated any tasks at all. Marketers need to look to AI and machine learning to automated remote, routine tasks to free up more time to create new campaigns. Source: Drift and Marketing Artificial Intelligence Institute, 2021 State of Marketing AI Report.

7.    Starbucks set the ambitious goal of being the world’s most personalized brand by relying on predictive analytics and machine learning to create a real-time personalization experience. The global coffee chain faced several challenges starting with how difficult it was to target individual customers with their existing IT infrastructure. They were also heavily reliant on manual operations across their thousands of stores, which made personalization at scale a formidable challenge to overcome. Starbucks created a real-time personalization engine that integrated with customers’ account information, the mobile app, customer preferences, 3^rd party data and contextual data. They achieved a 150% increase in user interaction using predictive analytics and AI, a 3X improvement in per-customer net incremental revenues. The following is a diagram of how DigitalBCG (Boston Consulting Group) was able to assist them. Source: Becoming The World’s Most Personalized Brand, DigitalBCG.  

8.    Getting personalization-at-scale right starts with a unified Customer Data Platform (CDP) that can use machine learning algorithms to discover new customer data patterns and “learn” over time.  For high-achieving marketing organizations, achieving personalization-at-scale is their highest and most urgent priority based on Salesforce Research’s most recent State of Marketing survey. And McKinsey predicts personalization-at-scale can create $1.7 trillion to $3 trillion in new value. For marketers to capture a part of this value, changes to the mar-tech stack (shown below) must be supported by clear accountability and ownership of channel and customer results. Combining a modified mar-tech stack with clear accountability delivers results.   Source: McKinsey & Company, A technology blueprint for personalization at scale. May 20, 2019. By Sean Flavin and Jason Heller.

9.    Campaign management, mobile app technology and testing/optimization are the leading three plans for a B2C company’s personalization technologies. Just 19% of enterprises have adopted AI and machine learning for B2C personalization today. The Forrester Study commissioned by IBM also found that 55% of enterprises believe the technology limitations inhibit their ability to execute personalization strategies. Source: A Forrester Consulting Thought Leadership Paper, Commissioned by IBM, Personalization Demystified: Enchant Your Customers By Going From Good To Great, February 2020.

10. Successful AI-driven personalization strategies deliver results beyond marketing, delivering strong results enterprise-wide, including lifting sales revenue, Net Promoter Scores and customer retention rates. When personalization-at-scale is done right, enterprises achieve a net 5.63% increase in sales revenue, 10.26% increase in order frequency, uplifts in average order value and an impressive 13.25% improvement in cross-sell/up-sell opportunities. The benefits transcend marketing alone and drive higher customer satisfaction metrics as well.   Source: A Forrester Consulting Thought Leadership Paper, Commissioned by IBM, Personalization Demystified: Enchant Your Customers By Going From Good To Great, February 2020.

CMOs and their teams rely on AI and machine learning to iteratively test and improve every aspect of their marketing campaigns and strategies. Striking the perfect balance between empathy and data-driven results takes a new level of data quality which isn’t possible to achieve using Microsoft Excel or personal productivity tools today. The most popular use of AI and machine learning in organizations is delivering personalization at scale across all digital channels. There’s also increasing adoption of predictive analytics based on machine learning to fine-tune propensity models to improve up-sell and cross-sell results. 

Bibliography

AI can help retailers understand the consumer, Phys.org. January 14, 2019

Brei, Vinicius. (2020). Machine Learning in Marketing: Overview, Learning Strategies, Applications and Future Developments. Foundations and Trends® in Marketing. 14. 173-236. 10.1561/1700000065.

Conick, H. (2017). The past, present and future of AI in marketing. Marketing News, 51(1), 26-35.

Drift and Marketing Artificial Intelligence Institute, 2021 State of Marketing AI Report.

Huang, M. H., & Rust, R. T. (2021). A strategic framework for artificial intelligence in marketing. Journal of the Academy of Marketing Science, 49(1), 30-50.

Jarek, K., & Mazurek, G. (2019). MARKETING AND ARTIFICIAL INTELLIGENCE. Central European Business Review, 8(2).

Libai, B., Bart, Y., Gensler, S., Hofacker, C. F., Kaplan, A., Kötterheinrich, K., & Kroll, E. B. (2020). Brave new world? On AI and the management of customer relationships. Journal of Interactive Marketing, 51, 44-56.

Ma, L., & Sun, B. (2020). Machine learning and AI in marketing–Connecting computing power to human insights. International Journal of Research in Marketing, 37(3), 481-504.

McKinsey & Company, A technology blueprint for personalization at scale. May 20, 2019

McKinsey Global Institute, Visualizing the uses and potential impact of AI and other analytics, April 17, 2018, | Interactive   

Microsoft Azure AI Gallery (https://gallery.azure.ai/)

Pedersen, C. L. Empathy‐based marketing. Psychology & Marketing.

Sinha, M., Healey, J., & Sengupta, T. (2020, July). Designing with AI for Digital Marketing. In Adjunct Publication of the 28th ACM Conference on User Modeling, Adaptation and Personalization (pp. 65-70).

State of Marketing, Sixth Edition. Salesforce Research, 2020.

The post 10 Ways AI And Machine Learning Are Improving Marketing In 2021 appeared first on Artificial Intelligence.

Yesterday, the team at PyTorch announced the availability of PyTorch Hub which is a simple API and workflow that offers the basic building blocks to improve machine learningresearch reproducibility.

Reproducibility plays an important role in research as it is an essential requirement for a lot of fields related to research including the ones based on machine learning techniques. But most of the machine learning based research publications are either not reproducible or are too difficult to reproduce.

With the increase in the number of research publications, tens of thousands of papers being hosted on arXiv and submissions to conferences, research reproducibility has now become even more important. Though most of the publications are accompanied by code and trained models that are useful but still it is difficult for users to figure out for most of the steps, themselves.

PyTorch Hub consists of a pre-trained model repository that is designed to facilitate research reproducibility and also to enable new research. It provides built-in support for Colab, integration with Papers With Code and also contains a set of models including classification and segmentation, transformers, generative, etc. By adding a simple hubconf.py file, it supports the publication of pre-trained models to a GitHub repository, which provides a list of models that are to be supported and a list of dependencies that are required for running the models.

For example, one can check out the torchvision, huggingface-bert and gan-model-zoorepositories. Considering the case of torchvision hubconf.py: In torchvision repository, each of the model files can function and can be executed independently. These model files don’t require any package except for PyTorch and they don’t need separate entry-points.

A hubconf.py can help users to send a pull request based on the template mentioned on the GitHub page.

The official blog post reads, “Our goal is to curate high-quality, easily-reproducible, maximally-beneficial models for research reproducibility. Hence, we may work with you to refine your pull request and in some cases reject some low-quality models to be published. Once we accept your pull request, your model will soon appear on Pytorch hub webpage for all users to explore.”

PyTorch Hub allows users to explore available models, load a model as well as understand the kind of methods available for any given model. Below mentioned are few of the examples:

Explore available entrypoints:

With the help of torch.hub.list() API, users can now list all available entrypoints in a repo.  PyTorch Hub also allows auxillary entrypoints apart from pretrained models such as bertTokenizer for preprocessing in the BERT models and making the user workflow smoother.

Load a model:

With the help of torch.hub.load() API, users can load a model entrypoint. This API can also provide useful information about instantiating the model.

Most of the users are happy about this news as they think it will be useful for them. A user commented on HackerNews, “I love that the tooling for ML experimentation is becoming more mature. Keeping track of hyperparameters, training/validation/test experiment test set manifests, code state, etc is both extremely crucial and extremely necessary.”

The post PyTorch announces the availability of PyTorch Hub for improving machine learning research reproducibility appeared first on Artificial Intelligence.