Deep Neural Networks (DNNs) have been successfully used in classifying digital images but have been less successful in classifying images with meanings that are not linear combinations of their visualized features, like images of artwork. Moreover, it is unknown what additional features must be included into DNNs, so that they can possibly classify using features beyond visually displayed features, like color, size, and form. Non-displayed features are important in abstract representations, reasoning, and understanding ambiguous expressions, which are arguably topics less studied by current AI methods. This paper attempts to identify capabilities that are related to semantic processing, a current limitation of DNNs. The proposed methodology identifies the missing capabilities by comparing the process of understanding Mondrian's paintings with the process of understanding electronic circuit designs, another creative problem solving instance. The compared entities are cognitive architectures that attempt to loosely mimic cognitive activities. The paper offers a detailed presentation of the characteristics of the architectural components, like goals, concepts, ideas, rules, procedures, beliefs, expectations, and outcomes. To explain the usefulness of the methodology, the paper discusses a new, three-step computational method to distinguish Mondrian's paintings from other artwork. The method includes in a backward order the cognitive architecture's components that operate only with the characteristics of the available data.

The ability to scale out training workloads has been one of the key performance enablers of deep learning. The main scaling approach is data-parallel GPU-based training, which has been boosted by hardware and software support for highly efficient point-to-point communication, and in particular via hardware bandwidth overprovisioning. Overprovisioning comes at a cost: there is an order of magnitude price difference between "cloud-grade" servers with such support, relative to their popular "consumer-grade" counterparts, although single server-grade and consumer-grade GPUs can have similar computational envelopes. In this paper, we show that the costly hardware overprovisioning approach can be supplanted via algorithmic and system design, and propose a framework called CGX, which provides efficient software support for compressed communication in ML applications, for both multi-GPU single-node training, as well as larger-scale multi-node training. CGX is based on two technical advances: \emph{At the system level}, it relies on a re-developed communication stack for ML frameworks, which provides flexible, highly-efficient support for compressed communication. \emph{At the application level}, it provides \emph{seamless, parameter-free} integration with popular frameworks, so that end-users do not have to modify training recipes, nor significant training code. This is complemented by a \emph{layer-wise adaptive compression} technique which dynamically balances compression gains with accuracy preservation. CGX integrates with popular ML frameworks, providing up to 3X speedups for multi-GPU nodes based on commodity hardware, and order-of-magnitude improvements in the multi-node setting, with negligible impact on accuracy.

Aligning users across networks using graph representation learning has been found effective where the alignment is accomplished in a low-dimensional embedding space. Yet, achieving highly precise alignment is still challenging, especially when nodes with long-range connectivity to the labeled anchors are encountered. To alleviate this limitation, we purposefully designed WL-Align which adopts a regularized representation learning framework to learn distinctive node representations. It extends the Weisfeiler-Lehman Isormorphism Test and learns the alignment in alternating phases of "across-network Weisfeiler-Lehman relabeling" and "proximity-preserving representation learning". The across-network Weisfeiler-Lehman relabeling is achieved through iterating the anchor-based label propagation and a similarity-based hashing to exploit the known anchors' connectivity to different nodes in an efficient and robust manner. The representation learning module preserves the second-order proximity within individual networks and is regularized by the across-network Weisfeiler-Lehman hash labels. Extensive experiments on real-world and synthetic datasets have demonstrated that our proposed WL-Align outperforms the state-of-the-art methods, achieving significant performance improvements in the "exact matching" scenario. Data and code of WL-Align are available at https://github.com/ChenPengGang/WLAlignCode.

You've seen the art AI image generators can create, and you may have played with natural language AI chatbots. You've benefited from artificial intelligence tools recommending you music and suggesting your next streaming show. But AI can do much more. Humans are excellent at spotting patterns. It's why we see faces on Mars or in the clouds.

Beachgoers could be safer thanks to a new technology with the potential to give real-time updates of rip currents. Rip currents are narrow, fast-moving segments of water that travel away from the shore. They can reach speeds of 2.5 meters per second, which is quicker than the fastest Olympic swimmer. NIWA and Surf Life Saving New Zealand (SLSNZ) have developed a state-of-the-art rip current identification tool using artificial intelligence (AI) and deep learning. The tool showed around 90% accuracy detecting rip currents in videos and images in trials.

This document presents endeavors to represent emotion in a computational cognitive architecture. The first part introduces research organizing with two axes of emotional affect: pleasantness and arousal. Following this basic of emotional components, the document discusses an aspect of emergent properties of emotion, showing interaction studies with human users. With these past author's studies, the document concludes that the advantage of the cognitive human-agent interaction approach is in representing human internal states and processes.

Most works on modeling the conversation history in Conversational Question Answering (CQA) report a single main result on a common CQA benchmark. While existing models show impressive results on CQA leaderboards, it remains unclear whether they are robust to shifts in setting (sometimes to more realistic ones), training data size (e.g. from large to small sets) and domain. In this work, we design and conduct the first large-scale robustness study of history modeling approaches for CQA. We find that high benchmark scores do not necessarily translate to strong robustness, and that various methods can perform extremely differently under different settings. Equipped with the insights from our study, we design a novel prompt-based history modeling approach, and demonstrate its strong robustness across various settings. Our approach is inspired by existing methods that highlight historic answers in the passage. However, instead of highlighting by modifying the passage token embeddings, we add textual prompts directly in the passage text. Our approach is simple, easy-to-plug into practically any model, and highly effective, thus we recommend it as a starting point for future model developers. We also hope that our study and insights will raise awareness to the importance of robustness-focused evaluation, in addition to obtaining high leaderboard scores, leading to better CQA systems.

During training, reinforcement learning systems interact with the world without considering the safety of their actions. When deployed into the real world, such systems can be dangerous and cause harm to their surroundings. Often, dangerous situations can be mitigated by defining a set of rules that the system should not violate under any conditions. For example, in robot navigation, one safety rule would be to avoid colliding with surrounding objects and people. In this work, we define safety rules in terms of the relationships between the agent and objects and use them to prevent reinforcement learning systems from performing potentially harmful actions. We propose a new safe epsilon-greedy algorithm that uses safety rules to override agents' actions if they are considered to be unsafe. In our experiments, we show that a safe epsilon-greedy policy significantly increases the safety of the agent during training, improves the learning efficiency resulting in much faster convergence, and achieves better performance than the base model.

The concept of the Metaverse aims to bring a fully-fledged extended reality environment to provide next generation applications and services. Development of the Metaverse is backed by many technologies, including, 5G, artificial intelligence, edge computing and extended reality. The advent of 6G is envisaged to mark a significant milestone in the development of the Metaverse, facilitating near-zero-latency, a plethora of new services and upgraded real-world infrastructure. This paper establishes the advantages of providing the Metaverse services over 6G along with an overview of the demanded technical requirements. The paper provides an insight to the concepts of the Metaverse and the envisaged technical capabilities of 6G mobile networks. Then, the technical aspects covering 6G for the development of the Metaverse, ranging from validating digital assets, interoperability, and efficient user interaction in the Metaverse to related security and privacy aspects are elaborated. Subsequently, the role of 6G technologies towards enabling the Metaverse, including artificial intelligence, blockchain, open radio access networks, edge computing, cloudification and internet of everything. The paper also presents 6G integration challenges and outlines ongoing projects towards developing the Metaverse technologies to facilitate the Metaverse applications and services.

Graph neural networks (GNNs) have demonstrated excellent performance in a wide range of applications. However, the enormous size of large-scale graphs hinders their applications under real-time inference scenarios. Although existing scalable GNNs leverage linear propagation to preprocess the features and accelerate the training and inference procedure, these methods still suffer from scalability issues when making inferences on unseen nodes, as the feature preprocessing requires the graph is known and fixed. To speed up the inference in the inductive setting, we propose a novel adaptive propagation order approach that generates the personalized propagation order for each node based on its topological information. This could successfully avoid the redundant computation of feature propagation. Moreover, the trade-off between accuracy and inference latency can be flexibly controlled by simple hyper-parameters to match different latency constraints of application scenarios. To compensate for the potential inference accuracy loss, we further propose Inception Distillation to exploit the multi scale reception information and improve the inference performance. Extensive experiments are conducted on four public datasets with different scales and characteristics, and the experimental results show that our proposed inference acceleration framework outperforms the SOTA graph inference acceleration baselines in terms of both accuracy and efficiency. In particular, the advantage of our proposed method is more significant on larger-scale datasets, and our framework achieves $75\times$ inference speedup on the largest Ogbn-products dataset.