The growing complexity of machine software as well as the ongoing modularization of modern production equipment has led to more simulation upfront. The fact that international travel for commissioning or service has significantly reduced or in some cases halted these days reinforces this trend. Functional tests of production equipment of the future will be performed using comprehensive models for simulation and virtual commissioning. The factory of the future will be built twice—first virtually, then physically. Digital representations of production machines continuously fed with live data from the field will be used for health monitoring throughout the entire lifetime of the equipment and will eventually make onsite missions be an exception ... Flexible production in the factory of the future will require robots and autonomous handling systems to adapt faster to changing requirements. While classic programming and teaching of robots isn’t suitable for preparing the system to handle the huge and fast-growing number of different goods, future handling equipment will automatically learn through reinforcement learning and other AI techniques. The prerequisites—massive calculation power and huge amounts of data—have been established over the past years.
With all of these security advantages, you might think that CISOs would have quickly moved to protect their applications and data by implementing secure enclaves. But market adoption has been limited by a number of factors. First, using the secure enclave protection hardware requires a different instruction set, and applications must be re-written and recompiled to work. Each of the different proprietary implementations of enclave-enabling technologies requires its own re-write. In most cases, enterprise IT organizations can’t afford to stop and port their applications, and they certainly can’t afford to port them to four different platforms. In the case of legacy or commercial off-the-shelf software, rewriting applications is not even an option. While secure enclave technologies do a great job protecting memory, they don’t cover storage and network communications – resources upon which most applications depend. Another limiting factor has been the lack of market awareness. Server vendors and cloud providers have quickly embraced the new technology, but most IT organizations still may not know about them.
Liquid networks make the model more robust by improving its resilience to unexpected and noisy data. For instance, it can make algorithms adjust to heavy rains that obscure a self-driving car’s vision. Liquid network makes the algorithm more interpretable. The network can help overcome the machine learning algorithms’ black-box nature because of the neurons’ expressive nature. The liquid network has performed better than other state-of-the-art time series by a few percentage points to predict future values in datasets used in atmospheric chemistry and traffic patterns. Apart from the high reliability, it also helped reduce computational costs. The researchers were aiming for fewer but richer nodes in the algorithm. In other words, the study focused on scaling down the network rather than scaling up. “This is a way forward for the future of robot control, natural language processing, video processing — any form of time series data processing,” said Ramin Hasani, the paper’s lead author. ... Tremendous progress has been made in developing smart bots that can perform multiple intelligent tasks like work alongside humans or give mental health advice. However, its adoption presents a significant concern in terms of safety and ethics.
The term cloud-native is still a large gray area and it's concept is still under discussion. If you, for example, read ten articles and books on the subject, all these materials will describe a different concept. However, what these concepts have in common is the same objective - get the most out of technologies within the cloud computing model. MicroProfile popularized this discussion and created a place for companies and communities to bring successful and unsuccessful cases. In addition, it promotes good practices with APIs, such as MicroProfile Config and the third factor of The Twelve-Factor App. ... The use of reflection by the frameworks has its trade-offs. For example, at the application start and in-memory consumption, the framework usually invokes the inner class ReflectionData within Class.java. It is instantiated as type SoftReference, which demands a certain time to leave the memory. So, I feel that in the future, some frameworks will generate metadata with reflection and other frameworks will generate this type of information at compile time like the Annotation Processing API or similar. We can see this kind of evolution already happening in CDI Lite, for example.
Overtime the classic PnP Sites Core has grown into a hard to maintain code base which made us decide to start a major upgrade effort for all PnP .NET components. As a result of that, PnP Framework is a slimmed down version of PnP Sites Core dropping legacy pieces and dropping support for on-premises SharePoint in favor of improved quality and maintainability. If you’re still using PnP Sites Core with your on-premises SharePoint than that’s perfectly fine, we’re not going to pull these components but you’ll not see any updated versions going forward. PnP Framework is a first milestone in the upgrade of the PnP .NET components, in parallel we’re building a brand new PnP Core SDK using modern .NET development techniques focused on performance and quality (check our test coverage and documentation). Overtime we’ll implement more and more of the PnP Framework functionality in PnP Core SDK and then replace the internal implementation in PnP Framework. The modern pages API is good example: when you use that API in PnP Framework you’re actually using the implementation done in PnP Core SDK. Below picture gives an overview of our journey and the road ahead:
While kernel mode is the most elevated type of access, it does come with several drawbacks that complicate EDR effectiveness. In kernel mode, visibility can be quite limited as there are several data points only available in user mode. Also, third-party kernel-based drivers are often difficult to develop and if not properly vetted can lead to higher chances of system instability. The kernel is often regarded as the most fragile part of a system and any panics or errors in kernel mode code can cause huge problems, even crashing the system entirely. User mode is often more appealing to attackers as it has no way of directly accessing the underlying hardware. Code that runs in user mode must use API functions that interact with the hardware on behalf of the application, allowing for more stability and fewer system-wide crashes (as application crashes will not affect the system). As a result, applications that run in user mode need minimal privileges and are more stable. Suffice to say, a lot of EDR products rely heavily on user mode hooks over kernel mode, making things interesting for attackers. Since the hooks exist in user mode and hook into our processes, we have control over them. Since applications run within the user’s context, this means everything that's loaded into our process can be manipulated by the user in some form or another.
For decades, enterprise software providers have focused on delivering large quarterly releases. "This system is slow because if there are any bugs in such a large release, developers have to sift through the deployed update in its entirety to find the problem to patch," said Eric Johnson, executive vice president of engineering for open-source code collaboration platform provider GitLab. Enterprises committed to CD rapidly deliver a string of highly granular releases. "This way, if there are any bugs in a new individual release they’re easily and swiftly addressed by developers' teams. Most developers appreciate CD because it helps them deliver higher quality work while limiting the risk of introducing unwanted change into production environments. CD ensures that the entire software delivery lifecycle from source control, to building and testing, to artifact release, and ultimately deployment into real environments, is automated and consistent, explained Brent Austin, director of engineering at Liberty Mutual Insurance. High levels of test automation are critical in CD, allowing developers to confidently introduce changes quickly with high confidence and higher quality. "CD also helps developers think in small batch sizes, which allows for easier and more effective rollback scenarios when issues are found and makes introducing change safer," Austin said.
Interacting with a ransomware operator is "unusual, but not that unusual," says Craig Williams, director of outreach for Cisco Talos. Of course, a key challenge in chatting with a criminal is knowing when to trust them. Researchers asked many questions they were able to verify, but there were scenarios in which they felt Aleks wasn't telling the whole story. Williams says the strongest example of this related to targeting the healthcare industry. "He pointed out how he didn't target healthcare customers … but then knew an awful lot about when healthcare paid, and in what situations they paid, and what type of data they have, and exactly how valuable it would be, and if they had insurance, they were more likely to pay," he explains. For example, Aleks reportedly told researchers hospitals pay 80% to 90% of the time. Aleks seems to choose victims based on their ability to pay quickly, Williams says, though the report notes the attacker's views may not represent those of LockBit group. For example, Aleks says the EU's General Data Protection Regulation (GDPR) may work in adversaries' favor. Victim companies are more likely to pay "quickly and quietly" so as to avoid penalties under GDPR.
As AI has bolstered the operations of more and more sectors, it’s become apparent that knowledge of the technology alone isn’t enough for deployments to succeed. Whether the AI solution is serving companies or individuals, the engineers behind the roll-out need to understand the business at hand. “The company needs people who know the principles of how these algorithms work, and how to train the machine, but can also understand the business domain and sector,” said Sanz-Saiz. Without this understanding, training an algorithm can be more complex. Any successful data scientist not only needs to bring technical expertise, but also needs to have domain and sector expertise as well.” Without sufficient industry knowledge, decision-making can become inaccurate, and in some cases, such as healthcare, it can also be dangerous. Companies such as Kheiron Medical have been using an AI solution to transform cancer screening, accelerating the process and minimising human error. For this to be effective, careful assessments and evaluations at every stage of the screening procedure need to be in place. “I think a commitment to clinical rigour needs to underpin everything that we do,” explained Sarah Kerruish, chief strategy officer at Kheiron.
AutoML is an automated process of searching for a child program from a search space to maximise a reward. The researchers broke down the process into a sequence of symbolic operations. Meaning, a child program is turned into a symbolic child program. The symbolic program is further hyperified into a search space by replacing some of the fixed parts with to-be-determined specifications. During the search, the search space materialises into different child programs based on search algorithm decisions. It can also be rewritten into a super-program to apply complex search algorithms such as efficient NAS (ENAS). PyGlove is a general symbolic programming library on Python. Using this library, Python classes, as well as functions, can be made mutable through brief Python annotations, making it easier to write AutoML programs. The library also allows AutoML techniques to be quickly dropped into preexisting machine learning pipelines while benefiting open-ended research which requires extreme flexibility. PyGlove implements various popular search algorithms, such as PPO, Regularised Evolution and Random Search.
Quote for the day:
"If you can't handle others' disapproval, then leadership isn't for you." -- Miles Anthony Smith