The Tinkerboard’s processor is more powerful than the one you’ll find in the Pi 4 B, so you may be able to get even more ambitious with your builds. However, when they’re available, you can get Pi 4s with up to 8 GB of RAM, which is more than the 2 GB that the Tinkerboard offers. Then there is the price. You can pick up a Tinkerboard S R2.0 on Amazon for $149.99 — which is more than some of the inflated Pi 4s are currently selling for. In short, this is a good option if you need more processing power or you can’t find a Pi 4, even at a premium. ... The Linux-powered ODROID XU4Q benefits from “Samsung Exynos5422 Cortex-A15 2Ghz and Cortex-A7 Octa core CPUs” along with 2GB of DDR3 RAM. On paper, this potentially makes the UX4Q the most powerful micro-computer on this list. It also comes with a very large heatsink attached, presumably to soak up some of the heat from its relatively powerful processor. With regards to ports, ODROID has managed to cram two USB 3.0, one USB 2.0, a Gigabit Ethernet, and an HDMI port onto the tiny board.
The human brain is hardwired to infer intentions behind words. Every time you engage in conversation, your mind automatically constructs a mental model of your conversation partner. You then use the words they say to fill in the model with that person’s goals, feelings and beliefs. The process of jumping from words to the mental model is seamless, getting triggered every time you receive a fully fledged sentence. This cognitive process saves you a lot of time and effort in everyday life, greatly facilitating your social interactions. However, in the case of AI systems, it misfires – building a mental model out of thin air. A little more probing can reveal the severity of this misfire. Consider the following prompt: “Peanut butter and feathers taste great together because___”. GPT-3 continued: “Peanut butter and feathers taste great together because they both have a nutty flavor. Peanut butter is also smooth and creamy, which helps to offset the feather’s texture.” The text in this case is as fluent as our example with pineapples, but this time the model is saying something decidedly less sensible.
AttacksResearchers are racing against hackers to develop stronger protections that keep data safe from malicious agents who would steal information by eavesdropping on smart devices. Much of the effort into preventing these “side-channel attacks” has focused on the vulnerability of digital processors. Hackers, for example, can measure the electric current drawn by a smartwatch’s CPU and use it to reconstruct secret data being processed, such as a password. MIT researchers recently published a paper in the IEEE Journal of Solid-State Circuits, which demonstrated that analog-to-digital converters in smart devices, which encode real-world signals from sensors into digital values that can be processed computationally, are vulnerable to power side-channel attacks. A hacker could measure the power supply current of the analog-to-digital converter and use machine learning algorithms to accurately reconstruct output data. Now, in two new research papers, engineers show that analog-to-digital converters are also susceptible to a stealthier form of side-channel attack, and describe techniques that effectively block both attacks.
It can be helpful to break apart the governance of AI algorithms into layers. At the lowest-level, close to the process are the rules of which humans have control over the training, retraining and deployment. The issues of accessibility and accountability are largely practical and implemented to prevent unknowns from changing the algorithm or its training set, perhaps maliciously. At the next level, there are questions about the enterprise that is running the AI algorithm. The corporate hierarchy that controls all actions of the corporation is naturally part of the AI governance because the curators of the AI fall into the normal reporting structure. Some companies are setting up special committees to consider ethical, legal and political aspects of governing the AI. Each entity also exists as part of a larger society. Many of the societal rule making bodies are turning their attention to AI algorithms. Some are simply industry-wide coalitions or committees. Some are local or national governments and others are nongovernmental organizations. All of these groups are often talking about passing laws or creating rules for how AI can be leashed.
For continuous operations to be successful, you must have infrastructure automation in place. In fact, continuous operations cannot exist without infrastructure automation. The true value that arises from the combination of infrastructure automation and continuous operations is that it gives back IT operations teams their time so they can focus on more complex reasoning or problem-solving tasks while the system simply continuously scans and fixes errors. ... The very essence of DevOps is constant change. Continuous operations may ultimately return your infrastructure to its desired state, but philosophically, it’s about being able to quickly and securely identify anomalies, apply fixes and modify your infrastructure as quickly as possible. It’s not as simple as flipping a switch or pushing a line of code. As the demand for security and compliance swells, continuous operations will have to build in these elements to be de facto checkboxes in the loop. At Puppet, we’ve baked continuous compliance and security into our infrastructure automation products to ensure continuous operations are indeed continuous.
A data culture creates standards for employee data literacy and provides open and transparent access to what assets exist, as well as standards for curation, quality, and certification so employees have a shared understanding of the data within an organization. “This will not resolve the silos, but it will create a transparent view of the entire enterprise data fabric,” Wills explains. He adds some of the approaches Alation has seen work well include things like providing an enterprise-wide data literacy training and certification program, so that everyone shares the same perspective, vocabulary, and basic analytic skills. Each functional business unit and area should include data training as part of their employee onboarding as it provides a review of an organization’s authoritative data and data-related assets, the process used to maintain them, and sets expectations for how employees should participate. “Also, recognition: Nothing motivates more and sends a stronger message than employees seeing each other be recognized and rewarded for their contributions,” Wills says.
One of the most obvious ways how digital real estate diverges from its physical counterpart is in the limited utility that it provides. This, of course, is because digital products do not require storage, nor do the digital people who populate the metaverse need to be kept comfortable or warm in indoor venues. However, the sense of discovery in the search for goods and services remains genuine within the metaverse, and it is in this way that virtual utility provides the most value. As businesses are free to design their purchased real estate however they want to, they can dedicate their efforts to creating the most eye-catching and exciting facades that will entice users to discover more about their property – and ultimately the goods and services they have on offer. Therefore, it is not so much about the utility of a piece of real estate that determines its valuation – but more about its network power. For example, how easy is it to discover this real estate? How well connected is it? What is the purchase power of the people coming to the piece of real estate? In this sense, valuing real estate in the metaverse, I’d argue, is a lot more like valuing a website, i.e., how many clicks does it get?
Quote for the day:
"Leadership - mobilization toward a common goal." -- Gary Wills