new One of the key benefits of AI tools in GRC is automation. GRC tasks can be time-consuming and complex, requiring businesses to stay up to date with changing regulations and laws. AI tools can help automate many of these tasks, ensuring that businesses are always compliant with regulations. For example, AI tools can help businesses create and update policies and procedures that ensure compliance with regulations. They can also help with monitoring compliance by analyzing data and identifying potential risks. — ISACA, in future

new In both cooking and machine learning, feedback is essential for improving the outcome. Chefs might taste a dish and adjust the seasoning, or ask for feedback from diners to improve their recipes. Similarly, machine learning algorithms rely on feedback to improve their accuracy and performance. In the case of supervised learning, the algorithm is trained on labeled data and receives feedback in the form of correct and incorrect predictions. In unsupervised learning, the algorithm receives feedback from the clustering or dimensionality reduction techniques used to analyze the data. — bbntimes.com, 4h ago

new Augmented intelligence and artificial intelligence are both concepts that are often used interchangeably. However, the difference between the two has important implications for our understanding and expectations of technology. Here, Dr. Gilad Wainreb, algorithms team leader at software company Lean AI, explains his understanding of the concept of augmented intelligence in the field of quality inspection and machine vision. — Metrology and Quality News - Online Magazine, 7h ago

new The ethical principles that relate to AI like Transparency, Accountability, Fairness, Security, Mitigating Bias and Privacy. — dsci.in, 5h ago

new ...10-20% of UK adults have experienced online abuse and, with large-scale public abuse part of daily life for many, it’s clear we need new and improved safety measures. Project-led by Jonathan Bright (The Alan Turing Institute), the Turing Online Safety team are using state-of-the-art language models to automatically detect abuse in social media data and produce responses. Explore the science behind these models and discover how data is created, with interactive language models and real-time results, at the team’s AI UK 2023 demo. — turing.ac.uk, 4h ago, Event

new The HPL-MxP benchmark seeks to highlight the convergence of HPC, and artificial intelligence (AI) workloads based on machine learning and deep learning by solving a system of linear equations using novel, mixed-precision algorithms that exploit modern hardware. — top500.org, 4h ago

new ...“All of the quantum computers currently available on Amazon Braket, the quantum computing service of AWS, started in the labs of experimental physicists. Innovation on these complex systems requires constant iteration on device design, fabrication methods, and control techniques. These devices require carefully isolated environments and delicate, complex components to facilitate interactions. The components themselves are often quite expensive, consisting of microwave, laser, and/or refrigeration technologies custom built in the lab or from boutique manufacturers. These factors all contribute to increase the resources required to build and experiment on quantum devices. — HPCwire, 5h ago

new Deep learning, part of machine learning, is based on multiple algorithms creating non-linear relationships in data. The growing use of deep learning algorithms in a wide variety of applications in agriculture, including crop and soil monitoring, insect and plant disease detection, and livestock health monitoring, is the major factor boosting the growth of the AI in agriculture market. — Future Farming, 8h ago

new In the context of AI safety, the precautionary principle can provide guidance for navigating the uncertainties and potential risks associated with AGI development. As AGI has the potential to surpass human intelligence and influence various aspects of our lives, ensuring its safety and alignment with human values is of utmost importance. Implementing the precautionary principle in AI development would involve:... — lesswrong.com, 19h ago, Event

new ...-- In quality assurance, it is often necessary to reliably detect deviations from the norm. Industrial cameras have a key role in this, capturing images of products and analyzing them for defects. If the error cases are not known in advance or are too diverse, however, rule-based image processing reaches its limits. By contrast, this challenge can be reliably solved with the AI method Anomaly Detection. The new, free IDS NXT 3.0 software update from IDS Imaging Development Systems makes the method available to all users of the AI vision system with immediate effect.The intelligent IDS NXT cameras are now able to detect anomalies independently and thereby optimize quality assurance processes. For this purpose, users train a neural network that is then executed on the programmable cameras. To achieve this, IDS offers the AI Vision Studio IDS NXT lighthouse, which is characterized by easy-to-use workflows and seamless integration into the IDS NXT ecosystem. Customers can even use only "GOOD" images for training. This means that relatively little training data is required compared to the other AI methods Object Detection and Classification. This simplifies the development of an AI vision application and is well suited for evaluating the potential of AI-based image processing for projects in the company.Another highlight of the release is the code reading function in the block-based editor. This enables IDS NXT cameras to locate, identify and read out different types of code and the required parameters. Attention maps in IDS NXT lighthouse also provide more transparency in the training process. They illustrate which areas in the image have an impact on classification results. In this way, users can identify and eliminate training errors before a neural network is deployed in the cameras.IDS NXT is a comprehensive AI-based vision system consisting of intelligent cameras plus software environment that covers the entire process from the creation to the execution of AI vision applications. The software tools make AI-based vision usable for different target groups – even without prior knowledge of artificial intelligence or application programming. In addition, expert tools enable open-platform programming, making IDS NXT cameras highly customizable and suitable for a wide range of applications.More information:... — PRLog, 19h ago

new Popular machine learning algorithms (such as neural networks, support vector machines and classification trees) operate within a probabilistic framework. They use probability to determine their optimal training and optimization parameters. — IoT Worlds, 1d ago

new The intelligent cameras are now able to detect anomalies independently and thereby optimise quality assurance processes. For this purpose, users train a neural network that is then executed on the programmable cameras. To achieve this, IDS offers the AI Vision Studio IDS NXT lighthouse, which is characterised by easy-to-use workflows and seamless integration into the IDS ecosystem. Customers can even use only ‘GOOD’ images for training. This means that relatively little training data is required compared to the other AI methods of object detection and classification. This simplifies the development of an AI vision application and is well-suited for evaluating the potential of AI-based image processing for projects in the company. — electronicspecifier.com, 19h ago

new In a new paper published in Neural Computation, Professor Terrence Sejnowski of the University of California San Diego and Salk Institute explores the relationship between the human interviewer and language models. Sejnowski suggests that language models reflect the intelligence and diversity of their interviewer, and that when he talks to ChatGPT, it seems as though another neuroscientist is talking back to him. This sparks larger questions about intelligence and what ‘artificial’ truly means. Sejnowski hopes to improve chatbot responses in the future, and his research provides insight into the relationship between humans and artificial intelligence. By understanding the intelligence and diversity of the interviewer, Sejnowski believes that chatbot responses can be improved and that the meaning of ‘artificial’ can be better understood. — Boxmining, 14h ago

new Building artificial systems that see and recognize the world similarly to human visual systems is a key goal of computer vision. Recent advancements in population brain activity measurement, along with improvements in the implementation and design of deep neural network models, have made it possible to directly compare the architectural features of artificial networks to those of biological brains’ latent representations, revealing crucial details about how these systems work. Reconstructing visual images from brain activity, such as that detected by functional magnetic resonance imaging (fMRI), is one of these applications. This is a fascinating but difficult problem because the underlying brain representations are largely unknown, and the sample size typically used for brain data is small. — MarkTechPost, 23h ago

new The Akitda platform has been developed by BrainChip, a company which develops edge AI on-chip processing and learning. Akida uses neuromorphic principles to mimic the human brain, analyzing only essential sensor inputs at the point of acquisition, and processing data with efficiency, precision, and economy of energy. The 2nd Generation of Akida can enhance vision, perception, and predictive applications in markets such as industrial and consumer IoT and personalized healthcare, etc. — Electronics For You, 1d ago

new To build on innovations that advance intelligence at the Edge, TI introduced a new family of six Arm Cortex-based vision processors that allow designers to add more vision and artificial intelligence (AI) processing at a lower cost, and with better energy efficiency, in applications such as video doorbells, machine vision and autonomous mobile robots. — electronicspecifier.com, 19h ago

new Enterprise automation so far has been mostly reactive, implemented as a piecemeal noninvasive method to automate routine, repetitive tasks, and structured processes and data. Business drivers, goals, and means for all the three vectors (Business processes, IT operations and Software development) of enterprise automation have expanded for the next chapter of the digital journey. Digital businesses need proactive, predictive end-to-end automation that leverages optimal blending of intelligent automation toolbox (e.g., process mining, conversational AI, machine learning and IDP) beyond RPA, supports human and machine synergies, and robust governance to accelerate strategic business innovation and unlock the success divide with competition. — IDC: The premier global market intelligence company, 19h ago, Event

new Deep learning deployment on the edge for real-time inference is key to many application areas. It significantly reduces the cost of communicating with the cloud in terms of network bandwidth, network latency, and power consumption. However, edge devices have limited memory, computing resources, and power. This means that a deep learning network must be optimized for embedded deployment. INT8 quantization has become a popular approach for such optimizations for ML frameworks like TensorFlow and PyTorch. SageMaker provides you with a bring your own container (BYOC) approach and integrated tools so that you can run quantization. — CoinGenius, 22h ago

new Accelerating the industry’s effort in this area, ipoque introduced encrypted traffic intelligence (ETI) across its suite of OEM DPI solutions, last year. “At its core, ETI features advanced AI-based analysis using ML, DL and high-dimensional data analysis. This includes ML / DL algorithms such as k-nearest neighbours (k-NN), decision tree learning models, convolutional neural networks (CNN), recurrent neural networks (RNN) and long short-term memory (LSTM) networks that boast over 6,000 features - including statistical, time series and packet-level features,” added Dr. Mieth. “We merge these with statistical and behavioural/heuristic analysis and DNS / service caching to accurately and reliably detect encrypted applications and services”. — electronicspecifier.com, 19h ago

new Arena enables the training and evaluation of embodied-AI models, along with the generation of new training data based on the human-robot interactions. It can thus contribute to the development of generalizable embodied agents with a wide variety of AI capabilities, such as task planning, visual dialogue, multimodal reasoning, task completion, teachable AI, and conversational understanding. — Amazon Science, 1d ago

new Training and running AI models is compute-intensive and costly – and does not compensate creators fairly for their work. Can decentralized technology provide a solution? Decentralized technology for AI models creation Many people think of all digital services and processes as essentially “free” in terms of energy consumption when compared with their real-world equivalents, but […]... — Outlier Ventures, 1d ago

new A traffic "Stop" sign on the roadside can be misinterpreted by a driverless vehicle as a speed limit sign when minimal graffiti is added. Wearing a pair of adversarial spectacles can fool facial recognition software into thinking that we are Brad Pitt. The vulnerability of artificial intelligence (AI) systems to such adversarial interventions raises questions around security and ethics, and many governments are now considering proposals for their regulation. I believe that mathematicians can contribute to this landscape. We can certainly get involved in the conflict escalation issue, where new defence strategies are needed to counter an increasingly sophisticated range of attacks. Perhaps more importantly, we also have the tools to address big picture questions, such as: What is the trade-off between robustness and accuracy? Can any AI system be fooled? Do proposed regulations make sense? Focussing on deep learning algorithms, I will describe how mathematical concepts can help us to understand and, where possible, ameliorate current limitations in AI technology. — ICMS - International Centre for Mathematical Sciences, 1d ago, Event

new Summary: Critics argue developers of generative AI systems such as ChatGPT and DALL-E have unfairly trained their models on copyrighted works. Those concerns are misguided. Moreover, restricting AI systems from training on legally accessed data would significantly curtail the development and adoption of generative AI across many sectors. Policymakers should focus on strengthening other IP rights to protect creators. (... — Center for Data Innovation, 1d ago

new With intuitive driver alerts and high-performance processing, the latest Driver•i dual-facing dash cameras “enable fleet management to promote safer driving, flag aggressive behavior, coach and communicate with drivers and exonerate them should accidents happen.”... — Insurtech Insights, 1d ago

new ...together with Geoffrey Hinton and Yann LeCun, said that current work on multimodal large neural nets, which have images or video as well as text, would “help a lot” with the “world model” issue — that is, that models need to understand the physics of our world. — VentureBeat, 1d ago

new The excellent performance of transformers in computer vision and natural language processing justifies research into the internal representations of these systems. Methods that involve training classifiers to infer latent features (such as part-of-speech and syntactic structure) are prevalent. — MarkTechPost, 1d ago

new From a compute perspective, AI everywhere requires a variety of processors to make it all happen: CPUs, GPUs, adaptable FPGAs and other accelerators across a range of capabilities, form factors and environmental tolerances. It requires chips for training the most complex of neural networks and chips for inferencing data in real time at the ruggedized edge. There’s chips for desktop AI, for AI in the cloud and AI in your handheld device. Putting together coherent AI deployments across this multi-layered landscape is an enormous challenge, stretching the very notion of heterogenous computing. — High-Performance Computing News Analysis | insideHPC, 1d ago

new The authors also warn against the use of AI in potentially harmful contexts like autonomous weapons or the manipulation of digital content for social destabilization, deplore the increasing centralization of decision-making power in the development of AI systems and biases embedded in them, and the lack of transparency and accountability in the industry. The book is downloadable... — Canadian Manufacturing, 1d ago

new ...“In this insightful discussion, the two participants explore the evolving relationship between AI and cybersecurity. The conversation covers a range of topics, from AI and machine learning (ML) tools in cyber defense and penetration testing to A/B testing in cyber attacks. The conversation also covers the challenges of AI and cybersecurity research and the maturity of AI-powered tools in the field. The conversation culminates in valuable advice for Chief Information Security Officers (CISOs) and Chief Information Officers (CIOs), as well as thoughts on the future of AI in cybersecurity, including the potential for cognitive attacks and the spread of misinformation.”... — The Tech Tribune, 1d ago

new To address these risks, you should establish strict guidelines for using ChatGPT, such as implementing access controls, data sanitization processes, and periodic reviews of AI-generated content. Additionally, you should ensure that AI-generated content is supplemented with human oversight and verification to guarantee accuracy and maintain client trust. By fostering a responsible and cautious approach to ChatGPT integration, MSPs and their vendors can better harness the potential benefits of AI while minimizing potential harm to their clients and the broader digital landscape. — The ChannelPro Network, 1d ago

new ...is a next-generation AI assistant based on Anthropic’s research into developing helpful, trustworthy, and harmless AI systems. Claude is available through their developer console’s chat interface and API and is capable of handling a wide range of conversational and text-processing tasks while still exhibiting a high level of dependability and predictability. — MarkTechPost, 1d ago

new Humans are capable of leaps of logic that machines are yet to catch up with. Basic computer programming is merely one level above AI. Recent advances and accomplishments in AI are indubitably tied to human intellectual capacity. Although machines are still capable of much more than the human brain, humans differ significantly in terms of application of their knowledge by using logic, reasoning, understanding, learning, and experience. — Analytics India Magazine, 1d ago

new There is an entire paradigm shift happening in the security and safety services market, mainly because of the adoption of IP-based cameras. Being connected to the Internet, these cameras allow remote surveillance. Moreover, because of the development of artificial intelligence (AI)-powered algorithms to decipher this data, the output is far more comprehensible than it has ever been before. — Zacks Investment Research, 1d ago

new With every company trying to inculcate the potential of AI ML into their services and product, MLOps has become popular. MLOps (Machine Learning Operations) is an essential function of Machine Learning engineering that mainly focuses on streamlining the process of putting ML models into production, followed by their maintenance and monitoring. It blends the features of both DevOps and ML to help organizations design robust ML pipelines with minimal resources and maximum efficiency. — MarkTechPost, 2d ago

new Another area of technology which promises to revolutionise the management of supply chains is artificial intelligence (AI), and its machine learning (ML) subset. Remember, it’s important to note the slight difference between AI and ML: AI enables a computer system to use maths and logic ‘think’ for itself and perform tasks autonomously. Meanwhile, ML allows the system to ‘learn’ and improve on its output, based on its experiences. — TechNative, 1d ago

new Critics argue developers of generative AI systems such as ChatGPT and DALL-E have unfairly trained their models on copyrighted works. Those concerns are misguided. — itif.org, 1d ago

new CMMs will be integrated with artificial intelligence and machine learning algorithms, allowing for the overall automation of quality inspection processes with improving accuracy and speed, while reducing the risk of human error. — Metrology and Quality News - Online Magazine, 1d ago

new The most valuable thing on the market is private data. AI developers, for example, need private data to improve the accuracy of their AI. — Our Bitcoin News, 1d ago

new Implementing transparency and explainability in AI models: This helps stakeholders understand how the AI system is making decisions, which could reveal instances of bias. — The European Business Review, 2d ago

new ...: Online e-learning platforms and intelligent tutoring systems (ITSs) provide convenient and controlled environments for learning, boosting accessibility and education standardization in and out of the classroom. For instance, many online ITSs are accessible to students 24/7 on desktops and mobile devices and provide automated feedback, maximizing learning robustness and schedule flexibility to students. A growing design consideration of ITSs is its customization to the student. Modern systems tailor the student experience through personalized content, assessments, remediation, and acceleration plans. Common challenges in these systems are that they collect minimal data and customization is slow, reactionary, and generalized. This talk discusses work that improves ITS effectiveness through real-time student performance modeling and machine learning to navigate these challenges. Specifically, the work explores personalized algorithmic solutions that predict when students need assistance and builds a digital student model framework for existing ITSs and e-learning platforms. — UCF Events, 1d ago

new ..., uses neuromorphic principles to mimic the human brain, analyzing only essential sensor inputs at the point of acquisition, processing data with unparalleled efficiency, precision, and economy of energy. Akida uniquely enables edge learning local to the chip, independent of the cloud, dramatically reducing latency while improving privacy and data security. Akida Neural processor IP, which can be integrated into SoCs on any process technology, has shown substantial benefits on today's workloads and networks, and offers a platform for developers to create, tune and run their models using standard AI workflows like Tensorflow/Keras. In enabling effective edge compute to be universally deployable across real world applications such as connected cars, consumer electronics, and industrial IoT, BrainChip is proving that on-chip AI, close to the sensor, is the future, for its customers' products, as well as the planet. Explore the benefits of Essential AI at... — tmcnet.com, 2d ago

new AI systems are the product of many different decisions made by those who develop and deploy them. From system purpose to how people interact with AI systems, we need to proactively guide these decisions toward more beneficial and equitable outcomes. That means keeping people and their goals at the center of system design decisions and respecting enduring values like fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability. — Inferse.com, 1d ago

new ...very large-scale language models with 100 billion parameters on logic-language understanding tasks. The team evaluated, for example, popular BERT pretrained language models with their “textual entailment” ones on stereotype, profession, and emotion bias tests. The latter outperformed other models with significantly lower bias, while preserving the language modeling ability. The “fairness” was evaluated with something called ideal context association (iCAT) tests, where higher iCAT scores mean fewer stereotypes. The model had higher than 90 percent iCAT scores, while other strong language understanding models ranged between 40 to 80. — Freethink, 1d ago, Event

new In this comprehensive guide, readers will gain a thorough understanding of AI topics, including deep learning, narrow AI, machine consciousness, robotics, and more. Patzer masterfully weaves complex concepts, real-world applications, and thought-provoking ethical discussions together. The book highlights the potential benefits of AI applications, such as improved accuracy and efficiency, as well as the risks and ethical considerations associated with its widespread implementation. — EIN Presswire, 2d ago