Articles

How Machine Learning Has Disrupted Finance Industry

,

In the last decade, the finance industry has seen an infusion of cutting-edge technologies like never before. This transformation is largely attributed to many startups that appeared on the scene post 2008 recession and followed a technology-first approach to create financial products and services with a target to improve customer experience. FinTech, as these startups are known, have been the early adopters of the new technologies like Smartphones, Big Data, Machine Learning (ML), Blockchain and were considered the trendsetters that were later followed by more traditional banks and financial institutes.

The recent advancements in machine learning and deep learning has really pushed the boundaries of computer vision and natural language processing. FinTechs are leaving no stones unturned to capitalize on these breakthroughs to improve financial services. As per a report, the ML Fintech market was valued at $7.27 billion in 2019 and it is expected to grow to $35.40 billion by 2025. Statista forecasts that the entire banking industry overall will be able to derive the business value of  $182 billion globally with machine learning by the year 2025.

Source de l’article sur DZONE

/par

Le fonds SAP.iO investit dans Jina AI, société de recherche basée sur les réseaux de neurones

, , ,

BERLIN – SAP SE (NYSE : SAP) annonce que son fonds d’investissement, SAP.iO Fund, a soutenu Jina AI, une entreprise berlinoise fournissant une solution de recherche basée sur les réseaux de neurones, en open source.

Jina AI combine les récentes avancées en matière de Machine Learning en traitement du langage naturel, vision et reconnaissance vocale dans une nouvelle plate-forme de recherche, afin d’offrir une plus grande précision, flexibilité et adaptabilité aux entrées de recherche.

Le projet principal de Jina AI, Jina on GitHub, permet aux utilisateurs de créer une solution de recherche native dans le cloud, fonctionnant sur la base du Deep Learning. Jina permet de réduire de plusieurs mois à quelques minutes le temps nécessaire à la construction d’un réseau de neurones prêt pour la production et adapté aux environnements commerciaux ; qui exigent un cycle de développement rapide. Depuis la sortie de On GitHub en mai 2020, ce projet a déjà attiré plus de 2 000 engagements de la part de 48 contributeurs du monde entier. Dès maintenant, Jina prend en charge la recherche de texte, d’image, de vidéo, d’audio et de données multimodales. D’autres types de données seront pris en charge à l’avenir.

« Alors que les entreprises accélèrent leurs transformations numériques, un besoin est apparu pour une recherche d’entreprise plus efficace et précise » a déclaré Ram Jambunathan, vice-président senior de SAP et directeur général de SAP.iO. « Nous sommes enthousiasmés par le potentiel de Jina AI à fournir une solution de recherche précise aux clients de SAP. »

Jina AI a été fondée par le Dr Han Xiao, qui est connu pour le développement de l’infrastructure de recherche de nouvelle génération pour l’application de messagerie de Tencent, WeChat. Il est également connu pour son leadership au sein du bureau du programme Open Source de Tencent, où il a encouragé la culture open source et de développement de la société. Xiao a été membre du conseil d’administration de la Linux Foundation AI en 2019 et est fondateur et président de l’association germano-chinoise de l’IA.

 

The post Le fonds SAP.iO investit dans Jina AI, société de recherche basée sur les réseaux de neurones appeared first on SAP France News.

Source de l’article sur sap.com

/par

7 Essential Tools for a Competent Data Scientist

,

A data scientist extracts manipulate and generate insights from humongous data. To leverage the power of data science, data scientists apply statistics, programming languages, data visualization, databases, etc.

So, when we observe the required skills for a data scientist in any job description, we understand that data science is mainly associated with Python, SQL, and R. The common skills and knowledge expected from a data scientist in the data science industry includes – Probability, Statistics, Calculus, Algebra, Programming, data visualization, machine learning, deep learning, and cloud computing. Also, they expect non-technical skills like business acumen, communication, and intellectual curiosity.

Source de l’article sur DZONE

/par

Eye Vs AI: Harnessing Technology to Successfully Track User Attention

Attention is the new gold; brands are in a constant competition for our attention.

A big portion of our time we spend online, where we are bombarded with insane amounts of information and advertisements. It’s hard not to become overwhelmed in this world of consumerism. We have had to become good at quickly evaluating which information is important, especially on the internet.

Good marketing specialists know that they have mere seconds to turn a potential customer into a lead. People are not going to spend a lot of time examining your advertisement or landing page, either it clicks or not. Moreover, most users do not read the articles, they scan them. First impression plays a huge role in the success of your business, so do not leave that to a chance.

You really don’t want your customer to ignore that special sale, subscription option, or another call to action on your webpage. That is why you need to know where that gold-worthy attention goes when a user opens your landing page. Here’s where technology can come in handy.

Eye-Tracking in Web Design

It is very important to know where your website visitor’s attention goes first. How to get that info? Eye-tracking is the answer.

Eye-tracking technology can be used to optimize your website conversions. By tracking eye movements, technology will recognize which content is most intriguing for the users. It will reveal whether people pay most attention where you want them to, which elements are distracting or not visible enough, and where sales are lost. This information is invaluable if you want to succeed in the current market.

This information is invaluable if you want to succeed in the current market

How does it work? An eye tracker, such as webcam or goggles, measures movement of an eye. Collected data is analyzed and presented as a heatmap, highlighting which elements of your design attract most attention. Having in mind that browsing time rarely exceeds a few seconds, this information is very valuable when you try to understand your audience.

You wouldn’t want to spend much time on your website design just to discover it does not generate desired conversion rate. By employing this technology you can make changes based on reliable data rather than intuition and guarantee your business future success.

By now you may think that you definitely need to carry out this eye-tracking study, but there is a catch. A high-quality behavioral observation or eye-tracking is a time-consuming, budget eating complicated process.

If you want to draw conclusions from heatmaps, you would need to include at least 39 participants in a study. One individual test may last from 20 minutes to an hour. Time quickly adds up when you include preparation and analysis of the results. The average eye tracker price is around $17,500 and it may vary between several thousand dollars and $50 000. Of course you can hire a company to carry out this research for you but it may cost you several hundred dollars a month. Luckily, technological innovations allow us to acquire the same insights about users’ attention flow much cheaper and faster than conducting or buying an actual eye-tracking study.

Technological innovations replace real eye-tracking study

AI-Powered Automatization of Eye-Tracking

In this task of understanding how internet users are interacting with your website, Artificial Intelligence (AI) seems to be an answer. AI-based technologies already have become prevalent in various services we use on a daily basis. For example, Netflix’s highly predictive algorithm offers viewers personalized movie recommendations. Medical researchers utilize eye tracking to diagnose conditions like Alzheimer’s disease or Autism. As these algorithms become better every year, AI also becomes an irreplaceable tool in business.

Over the years researchers have collected so much data that human behavior becomes really predictable

How can AI help you to understand your customer’s attention? The main feature of AI is that it can mimic human intelligence and constantly improve itself by learning from data. Predictive eye-tracking is based on deep learning and trained with previous eye tracking study data. Over the years researchers have collected so much data that human behavior becomes really predictable. Technology predicts which specific areas of your website attract most interest. In this way, AI enables you to speed up the UX research process and get insights about your design in a matter of seconds.

Too good to be true? There are already several available tools on the market, such as Attention Insight or EyeQuant. These predictive design tools are based on deep learning and trained with previous eye-tracking studies data. Up to date, they have achieved an 84-90% accuracy.

AI-powered attention heatmap

AI solutions for designers and marketers have already become major competitors to traditional eye-tracking studies. Due to active competition, predictive eye-tracking tools are constantly innovating and recently started generating heatmaps for videos. Another useful feature that provides decision-makers with quantitative data is a percentage of attention. Users can define an object that they want to test and get an exact percentage of attention that the object receives.

Conclusion

Since all digital products are competing for user’s limited attention, it has become one of the most valuable resources. Due to fierce competition, it is not enough to rely on your intuition and gut instinct while making important decisions anymore. Designers have a choice in this economy of attention, though.

Yes, there are eye-tracking studies that require a significant amount of time and financial resources.

However, you can make user-centric, data-driven decisions in a quick, scalable, and private way while your product is still under development. AI-powered predictive eye-tracking tools might be an answer. Attention is a new currency, and you must measure it.

Source

p img {display:inline-block; margin-right:10px;}
.alignleft {float:left;}
p.showcase {clear:both;}
body#browserfriendly p, body#podcast p, div#emailbody p{margin:0;}

Source de l’article sur

/par

2 Things You Need to Know About Reinforcement Learning: Computational and Sample Efficiency

,

The High Cost of Deep Learning

Have you ever put on a sweater because the air conditioning was too cold? Forgotten to turn off the lights in another room before heading to bed? Do you commute to work more than 30 minutes every day just for the sake of “filling seats” at the office, even though everything you do at work could be done via laptop from home? 

In the counter-intuitive trade-offs between sample and computational efficiency in Reinforcement Learning, choosing evolution strategies can be smarter than it looks.

Source de l’article sur DZONE

/par

Deep Learning for Manufacturing: Overview and Applications

,
Deep learning

Introduction to Deep Learning for Manufacturing

Before getting into the details of deep learning for manufacturing, it’s good to step back and view a brief history. Concepts, original thinking, and physical inventions have been shaping the world economy and manufacturing industry since the beginning of the modern era, i.e. early 18th century.

Ideas of economies-of-scale by the likes of Adam Smith and John Stuart Mill, the first industrial revolution and steam-powered machines, electrification of factories and the second industrial revolution, and the introduction of the assembly line method by Henry Ford are just some of the prime examples of how the search for high efficiency and enhanced productivity have always been at the heart of manufacturing.

Source de l’article sur DZONE

/par

What You Must Know Before You Dive Into Machine Learning

,

Machine learning refers to the process of enabling computer systems to learn with data using statistical techniques without being explicitly programmed. It is the process of active engagement with algorithms in order to enable them to learn from and make predictions on data. Machine learning is closely associated with computational statistics, mathematical optimization, and data learning. It is associated with predictive analysis, which allows producing reliable and fast results by learning from historical trends. There are basically two kinds of machine learning tasks:

  1. Supervised learning: The computer is presented with some example inputs, based on which the desired outputs are to be formed. The computer is made to learn general rules of converting inputs to outputs.

    Source de l’article sur DZONE

/par

Microsoft Launches Latest Bid to One Up AI Competition

,

Photo credit by Unsplash/Hermes Rivera

It’s no secret these days that AI is a big deal in the world of business. According to Gartner, the percentage of enterprises using the technology has jumped astronomically over the past several years, tripling in the last year alone.

Source de l’article sur DZONE

/0 Commentaires/par

Comparing bfloat16 Range and Precision to Other 16-bit Numbers

, ,

Deep Learning has spurred interest in novel floating point formats. Algorithms often don’t need as much precision as standard IEEE-754 doubles or even single precision floats. Lower precision makes it possible to hold more numbers in memory, reducing the time spent swapping numbers in and out of memory. Since this where a lot of time goes, low precision formats can speed things up quite a bit.

Here I want to look at bfloat16, or BF16 for short, and compare it to 16-bit number formats I’ve written about previously, IEEE and posit.


Source de l’article sur DZONE (AI)

/0 Commentaires/par