Connect with us

AI

The Third Pillar of Trusted AI: Ethics

Published

on

Click to learn more about author Scott Reed.

Building an accurate, fast, and performant model founded upon strong Data Quality standards is no easy task. Taking the model into production with governance workflows and monitoring for sustainability is even more challenging. Finally, ensuring the model is explainable, transparent, and fair based on your organization’s ethics and values is the most difficult aspect of trusted AI.

We have identified three pillars of trust: performance, operations, and ethics. In our previous articles, we covered performance and operations. In this article, we will look at our third and final pillar of trust, ethics.

Ethics relates to the question: “How well does my model align with my organization’s ethics and values?” This pillar primarily focuses on understanding and explaining the mystique of model predictions, as well as identifying and neutralizing any hidden sources of bias. There are four primary components to ethics: 

  • Privacy
  • Bias and fairness
  • Explainability and transparency
  • Impact on the organization

In this article, we will focus on two in particular: bias and fairness and explainability and transparency. 

Bias and Fairness

Examples of algorithmic bias are everywhere today, oftentimes relating to the protected attributes of gender or race, and existing across almost every vertical, including health care, housing, and human resources. As AI becomes more prevalent and accepted in society, the number of incidents of AI bias will only increase without standardized responsible AI practices.

Let’s define bias and fairness before moving on. Bias refers to situations in which,  mathematically, the model performed differently (better or worse) for distinct groups in the data. Fairness, on the other hand, is a social construct and subjective based on stakeholders, legal regulations, or values. The intersection between the two lies in context and the interpretation of test results.

At the highest level, measuring bias can be split into two categories: fairness by representation and fairness by error. The former means measuring fairness based on the model’s predictions among all groups, while the latter means measuring fairness based on the model’s error rate among all groups. The idea is to know if the model is predicting favorable outcomes at a significantly higher rate for a particular group in fairness by representation, or if the model is wrong more often for a particular group in fairness by error. Within these two families, there are individual metrics that can be applied. Let’s look at a couple of examples to demonstrate this point.

In a hiring use case where we are predicting if an applicant will be hired or not, we would measure bias within a protected attribute such as gender. In this case, we may use a metric like proportional parity, which satisfies fairness by representation by requiring each group to receive the same percentage of favorable predictions (i.e., the model predicts “hired” 50% of the time for both males and females). 

Next, consider a medical diagnosis use case for a life-threatening disease. This time, we may use a metric like favorable predictive value parity, which satisfies fairness by equal error by requiring each group to have the same precision, or probability of the model being correct. 

Once bias is identified, there are several different ways to mitigate and force the model to be fair. Initially, you can analyze your underlying data, and determine if there are any steps in data curation or feature engineering that may assist. However, if a more algorithmic approach is required, there are a variety of techniques that have emerged to assist. At a high level, those techniques can be classified by the stage of the machine learning pipeline in which they are applied:

  • Pre-processing
  • In-processing
  • Post-processing

Pre-processing mitigation happens before any modeling takes place, directly on the training data. In-processing techniques relate to actions taken during the modeling process (i.e., training). Finally, post-processing techniques occur after modeling the process and operate on the model predictions to mitigate bias.

Explainability and Transparency

All Data Science practitioners have been in a meeting where they were caught off-guard trying to explain the inner workings of a model or the model’s predictions. From experience, I know that isn’t a pleasant feeling, but those stakeholders had a point. Trust in ethics also means being able to interpret, or explain, the model and its results as well as possible. 

Explainability should be a part of the conversation when selecting which model to put into production. Choosing a more explainable model is a great way to build rapport between the model and all stakeholders. Certain models are more easily explainable and transparent than others – for example, models that use coefficients (i.e., linear regression) or ones that are tree-based (i.e., random forest). These are very different from deep learning models, which are far less intuitive. The question becomes, should we sacrifice a bit of model performance for a model that we can explain?

At the model prediction level, we can leverage explanation techniques like XEMP or SHAP to understand why a particular prediction was assigned to the favorable or unfavorable outcome. Both methods are able to show which features contribute most, in a negative or positive way, to an individual prediction. 

Conclusion

In this series, we have covered the three pillars of trust in AI: performance, operations, and ethics. Each plays a significant role in the lifecycle of an AI project. While we’ve covered them in separate articles, in order to fully trust an AI system, there are no trade-offs between the pillars. Enacting trusted AI requires buy-in at all levels and a commitment to each of these pillars. It won’t be an easy journey, but it is a necessity if we want to ensure the maximum benefit and minimize the potential for harm through AI. 

Coinsmart. Beste Bitcoin-Börse in Europa
Source: https://www.dataversity.net/the-third-pillar-of-trusted-ai-ethics/

Artificial Intelligence

Same-day delivery apps need more than speed to survive post-pandemic

Published

on

We have entered a whole new era of e-commerce centered on speed and convenience. Business leaders are being forced to prioritize delivery capabilities and push for more accelerated delivery services.

“Fast/reliable delivery” was the most important online shopping attribute among the more than 8,500 consumers queried for PwC’s June 2021 Global Consumer Insights Pulse Survey, making it clear that delivery services will only become more crucial across the e-commerce landscape.

Now that consumers have grown accustomed to same-day (and same-hour) delivery service models, customer expectations for delivery options will only increase.

In fact, according to a recent report from the mobile app intelligence platform SensorTower, the top food delivery apps saw continued growth in January and February 2021, with installs up 14% year over year. And yet, despite climbing user growth, DoorDash, Uber Eats and GrubHub remain unprofitable. So how can business leaders design rapid delivery models that meet consumer expectations — and still make money?

If your delivery service results in a poor customer experience, you’ll be less likely to win customer loyalty just because you offer faster delivery.

The challenge: Delivery apps need more than speed to drive profitability

To remain competitive, delivery apps are rethinking their services and broadening their offerings.

“Amazon powers next-day delivery,” Raj Beri, Uber’s global head of grocery and new verticals, said in May. “We’re going to power next-hour commerce.”

But speeding up the delivery process won’t necessarily drive revenue. More importantly, if your delivery service results in a poor customer experience, you’ll be less likely to win customer loyalty just because you offer faster delivery.

The primary challenge faced by delivery apps, or any e-commerce company looking to add delivery services as part of its offerings, is building a foundation that enables not only speed and convenience for the customer, but one that takes into account all aspects of the customer experience. For example, when delivering food, the business responsible for the delivery must make sure the food is handled safely and remain free of any contaminants. The temperature — whether hot or cold — must be maintained throughout the delivery process and the order itself must be correct.

The solution: Same-day delivery relies on sophisticated technology platforms

The “Uberization” of everything, combined with dramatically elevated consumer expectations, will take much more than a delivery app and fleet of drivers for businesses to be profitable. To follow through on the promise of same-day delivery services, a number of things need to happen without any missteps between when an order is placed and when it shows up at the customer’s door. The more complex the product being delivered, the more difficult the delivery process becomes.

To enable same-day delivery services while also reaching profitability, a delivery app must take into account the technology needed to meet customer expectations. It involves much more than simply designing an app and growing user numbers. A truly successful same-day delivery model that provides an exceptional customer experience relies on a sophisticated software platform that can simultaneously manage various aspects of the customer journey, all while making it appear seamless from the customer’s point of view.

Profitable delivery services are built on automated systems powered by artificial intelligence systems and robotics. The technology must come first, before the app and before user growth. Any other delivery business model is putting the cart before the horse.

Domino’s Pizza is a brand that has perfected the delivery process and vastly improved the overall customer experience by making technology core to their business model. The key moment came when the brand defined itself as an e-commerce company that sells pizza. It committed to data applications and implemented a robotics technology platform that enabled electronic delivery systems that added speed and efficiency to the delivery process. In April, Domino’s began rolling out a robot car delivery service to select customers in Houston via Nuro.

GrubHub is also taking steps to integrate robotic capabilities into its delivery process. According to recent reports, the company announced it would be adding self-driving units that deploy drone-like robots to deliver food to college students. The program, which will roll out on a limited number of U.S. college campuses this fall, aims to reduce delivery times and, hopefully, costs.

This focus on technology is crucial in the world of delivery apps, or for any businesses forced to compete in the newly emerging category of next-hour commerce. The key to building a successful, profitable business model is to invest in technology platforms that can connect all components of the customer journey, from opening an app and clicking on a product to purchasing the product and scheduling the delivery, and beyond.

Same-day delivery: Where to go from here

In a world where everyone wants to open an app on their phone and have whatever it is they need to be delivered within an hour, it’s tempting for business leaders to focus on the delivery app itself, whether they are building their own or partnering with another company. But focusing on the app is a shortsighted view of same-day delivery models.

Instead, business leaders must use a wide-angle lens and consider every single aspect of their customer journey: How do customers engage with their business? How do customers search for and find the products they offer? What does it take to complete an order and what conditions must be met before the order can be delivered? Also, what happens after the order to ensure it went smoothly and to the customer’s satisfaction?

Some businesses are finding success partnering with delivery apps, but this comes with the risk of putting your brand’s reputation in the hands of another company that acts as a frontline employee with customers. Other companies are adding delivery service options to their current e-commerce model, relying on third-party software that can be plugged into an existing technology stack. Unfortunately, this comes with limitations and is not viable for regulated businesses that include multiple components.

The only way to ensure a seamless customer experience on top of same-day delivery services is to build a proprietary software platform that puts the technology at the heart of your business, which allows you to automate key processes, adding speed and convenience to your delivery model. It also makes it possible to integrate robotic systems that can expedite orders, include artificial intelligence protocols that can accelerate business growth, and scale your delivery model as your business expands.

Thriving in the new era of e-commerce

“Next-hour delivery” is a catchy tagline that is sure to gain traction among consumers, but whether it will help drive profitability remains to be seen. As the CEO of a firm that has built a profitable business model centered on same-day delivery services, I’m skeptical that the promise of next-hour delivery will drive more revenue if the technology powering the delivery systems lacks automation, artificial intelligence and robotics.

It’s true that businesses will be forced to compete on same-day delivery. But another truth that has emerged since the pandemic is that this new era of e-commerce comes with heightened customer expectations that won’t be met on speed alone. Consumer satisfaction hinges on more than the amount of time it takes to move an order from an app to the customer’s door.

To succeed in the delivery service market, business leaders must ask themselves a number of questions: Which parts of their business are needed to complete a same-day delivery order? Is the ordering process intuitive? Can the order and delivery be monitored by the customer? Is the order correct when it arrives? Does it meet the customer’s expectations?

And, most importantly, is their business built on a technology platform that can support the entire customer journey and delivery model, from product discovery and purchase to same-day delivery and beyond? The businesses that answer yes to these questions are the ones I expect to thrive in the post-pandemic world.

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://techcrunch.com/2021/07/27/same-day-delivery-apps-need-more-than-speed-to-survive-post-pandemic/

Continue Reading

AI

ARTIFICIAL INTELLIGENCE (AI), A TEXTBOOK

Published

on

ARTIFICIAL INTELLIGENCE (AI), A TEXTBOOK

This book covers the broader field of AI, carefully balancing coverage between classical AI (logic or deductive reasoning) and modern AI (inductive learning and neural networks).


Sponsored Post.

Artificial Intelligence: A Textbook (Springer),
by Charu C. Aggarwal, June 2021

.
Charu Aggarwal AI Textbook
Table of Contents

PDF Download Link (Free for computers connected to subscribing institutions only)

Buy hardcover from Springer or Amazon (for general public)

Buy low-cost paperback edition (MyCopy link on right appears only for computers connected to subscribing institutions)

This book covers the broader field of artificial intelligence. The book carefully balances coverage between classical AI (logic or deductive reasoning) and modern AI (inductive learning and neural networks). The chapters of this book span three categories:

Deductive reasoning methods:
These methods start with pre-defined hypotheses and reason with them in order to arrive at logically sound conclusions. The underlying methods include search and logic-based methods. These methods are discussed in Chapters 1 through 5.

Inductive learning methods:
These methods start with examples and use statistical methods in order to arrive at hypotheses. Examples include regression modeling, support vector machines, neural networks, reinforcement learning, unsupervised learning, and probabilistic graphical models. These methods are discussed in Chapters 6 through 11.

Integrating reasoning and learning:
Chapters 12 and 13 discuss techniques for integrating reasoning and learning. Examples include the use of knowledge graphs and neuro-symbolic artificial intelligence.

The book is available in both hardcopy (hardcover) and electronic versions.

The hardcover is available at all the usual channels (e.g, Amazon, Barnes and Noble etc.), in Kindle format, and also directly from Springer in hardcopy and pdf format. PDF versions do have links and work with e-readers (including the kindle reader). The PDF version (bought directly from Springer) provides better formatting of equations than the kindle version and has an almost identical layout and pagination to the hardcopy on the e-reader.


PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://www.kdnuggets.com/2021/07/charu-ai-textbook.html

Continue Reading

AI

Cupix digital twin plugs into Autodesk BIM 360 for 3D builder workflows

Published

on

All the sessions from Transform 2021 are available on-demand now. Watch now.


Construction digital twins pioneer Cupix today announced an integration with Autodesk’s BIM 360 construction management platform. This is intended to streamline construction workflows that weave up-to-date information about the construction process into Autodesk planning tools.

Cupix’s move builds on a prior integration into the Autodesk PlanGrid platform for construction planning. For the vendor and its customers alike, such integrations with the Autodesk environment are a key to bringing digital twins to wider markets. As a mainstay provider of tools for organizing architectural, engineering, and construction management processes, Autodesk will likely influence uptake of digital twins in these key sectors.

“We believe the 3D digital twin platform will come to be seen as a new IT pillar — in the same way ERP, BIM, CRM, and groupware are relied on to improve corporate-wide productivity,” Cupix CEO and Founder Simon Bae told VentureBeat.

He said Cupix’s goal is to simplify the process of capturing real-time data about construction progress using low-cost cameras. This allows remote contractors, managers, owners, and architects to virtually walk through job sites, create new requests for information (RFIs), and assign them based on what they see and learn in the virtual walkthrough.

Streamlining digital twin workflows

Cupix’s special sauce lies in reducing the time, cost, and effort needed to capture up-to-date spatial data in a 3D digital twin for construction. This complements other tools that generate 3D walkthroughs from drawings.

“To date, construction remains largely a 2D industry and one that hungers for technological innovation,” Bae said. “We believe 3D digital twin technology, and CupixWorks in particular, is a game-changer for customers.”

Importantly, Cupix allows non-technical users to capture a 3D representation of a job site using a consumer-grade 360-degree camera, rather than high-end cameras or lidar, components that can cost tens of thousands of dollars. With the Cupix approach, teams can update scans daily rather than waiting days or weeks for a fresh scan.

Benefits go beyond the act of data capture because traditional 3D scan data eats up a lot of bytes.

“You can easily end up with several gigabytes of data after scanning just 10,000 square feet of space,” Bae said. Cupix has focused on reducing data requirements while preserving enough fidelity for everyday use cases, he indicated.

Cupix has particularly focused on improving user experience and workflows in the construction industry. Bae argues that other 3D scanning platforms, such as Matterport, focus on wider sectors, with different requirements. Although they may provide high-resolution imagery at a low price, it can be time-consuming to complete regular 3D scans of an actual job site, making them less useful when it comes to frequently capturing data on job site views during construction, Bae maintains.

“We believe that the Cupix approach will deliver to customers the collaboration, confidence, and control they need to stay on time, on budget, and on target,” Bae said. That is important if digital twin technology is going to fulfill its promise of bringing digital transformation to construction.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact. Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://venturebeat.com/2021/07/27/cupix-digital-twin-plugs-into-autodesk-bim-360-for-3d-builder-workflows/

Continue Reading

AI

Regulatory Risk Faced by Apple’s App Store

Published

on

Robust sales of 5G iPhones and services such as the Apple Music and App Store are expected to grow at Apple Inc (AAPL.O). The service business is subject to litigation from the U.S. Department of Justice. Also, pending legislation in the United States and Europe will reduce their commissions on applications and make further changes.

“We believe that government action (with antitrust, executive order, and legislation) represents one major risk to Apple’s shares,” said Tom Forte, an analyst at D. Davidson & Co, who wrote the letter to clients and added that he hoped Apple’s management would address the risks posed by the company’s revenue today.

Effect on Sales

Analysts expect Apple’s service sales to extend by 24.1% to $ 16.33 billion, almost a fifth of the entire sales revenue of $ 73.30 billion, consistent with IBES data from Refinitiv from July 26.

Epic Games sued Apple for its commissions on the App Store in 2020. Cowen & Co analyst Krish Sankar estimated that this happened because the Apple App Store was offering about 6% of its total revenue and out of 10 % and 15% of its profit.

Risks Faced by Apple

Apple also faces the risk that the U.S. Department of Justice can bar Google’s Alphabet Inc (GOOGL.O) from paying as a search engine on the iPhone, Angelo Zino of CFRA Research wrote in a research book last week. Officials from the Department of Justice have pointed out the limit on Google paying Apple $ 8 billion to $ 12 billion a year.

Projection of iPhone Sales

Meanwhile, analysts expect the company to see an increase in iPhone sales in its third-quarter financially, with sales rising by 28.7% to $ 34 billion, according to Refinitiv data from July 26, the largest contributor to sales.

JP Morgan analyst Samik Chatterjee wrote in a letter last week that the growth of Apple’s business led to a turnaround in the company’s high value, with shares trading at around $ 30. As recently as 2019, stocks have been trading regularly for less than 20 times as investors have always been cautious about the company’s high reliance on iPhone sales.

Recommended Products

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://1reddrop.com/2021/07/27/regulatory-risk-faced-by-apples-app-store/?utm_source=rss&utm_medium=rss&utm_campaign=regulatory-risk-faced-by-apples-app-store

Continue Reading
Esports4 days ago

Teppei Genshin Impact Voice Actor: Who is it?

Esports4 days ago

Who won Minecraft Championships (MCC) 15? | Final Standings and Scores

Esports5 days ago

Can You Play Pokemon UNITE Offline?

Esports4 days ago

All ranked mode rewards for Pokémon UNITE: Season 1

Aviation3 days ago

Legendary F-14 Pilot Dale ‘Snort’ Snodgrass Dies In A Tragic Plane Crash

Cleantech4 days ago

Form Energy Reveals Iron-Air 100 Hour Storage Battery

Esports4 days ago

Sakura Arborism Genshin Impact: How to Complete

Esports4 days ago

Here are the results for the PUBG Mobile World Invitational (PMWI) West 2021

watch-live-russias-pirs-module-set-to-depart-space-station-today.jpg
Aerospace2 days ago

Watch live: Russia’s Pirs module set to depart space station today

Esports4 days ago

Here are the results for the PUBG Mobile World Invitational (PMWI) East 2021

Techcrunch4 days ago

This Week in Apps: Clubhouse opens up, Twitter talks bitcoin, Snap sees record quarter

best-gengar-build-in-pokemon-unite.png
Esports4 days ago

Best Gengar build in Pokémon UNITE

Cyber Security4 days ago

Threat Actors are Abusing Argo Workflows to Target Kubernetes

Esports4 days ago

Are there ranked rewards in Pokémon UNITE?

Esports4 days ago

Best Garchomp build in Pokémon UNITE

Cyber Security4 days ago

What Programming Language Should I Learn for CyberSecurity?

Esports3 days ago

How to unlock Pokémon in Pokémon UNITE, all Unite License costs

AR/VR4 days ago

Warplanes: WW1 Fighters to See Official Oculus Quest Store Launch This Week

AI3 days ago

What is the Freedom Phone and Should You Buy It?

Blockchain4 days ago

Canadian Border Town Halts Crypto Mining to Draw Up Regulations

Trending