Medical image segmentation is inherently uncertain. For a given image, there may be multiple plausible segmentation hypotheses, and physicians will often disagree on lesion and organ boundaries. To be suited to real-world application, automatic segmentation systems must be able to capture this uncertainty and variability. Thus far, this has been addressed by building deep learning models that, through dropout, multiple heads, or variational inference, can produce a set - infinite, in some cases - of plausible segmentation hypotheses for any given image. However, in clinical practice, it may not be practical to browse all hypotheses. Furthermore, recent work shows that segmentation variability plateaus after a certain number of independent annotations, suggesting that a large enough group of physicians may be able to represent the whole space of possible segmentations. Inspired by this, we propose a simple method to obtain soft labels from the annotations of multiple physicians and train models that, for each image, produce a single well-calibrated output that can be thresholded at multiple confidence levels, according to each application's precision-recall requirements. We evaluated our method on the MICCAI 2021 QUBIQ challenge, showing that it performs well across multiple medical image segmentation tasks, produces well-calibrated predictions, and, on average, performs better at matching physicians' predictions than other physicians.

Collaborative filtering (CF) is a widely studied research topic in recommender systems. The learning of a CF model generally depends on three major components, namely interaction encoder, loss function, and negative sampling. While many existing studies focus on the design of more powerful interaction encoders, the impacts of loss functions and negative sampling ratios have not yet been well explored. In this work, we show that the choice of loss function as well as negative sampling ratio is equivalently important. More specifically, we propose the cosine contrastive loss (CCL) and further incorporate it to a simple unified CF model, dubbed SimpleX. Extensive experiments have been conducted on 11 benchmark datasets and compared with 29 existing CF models in total. Surprisingly, the results show that, under our CCL loss and a large negative sampling ratio, SimpleX can surpass most sophisticated state-of-the-art models by a large margin (e.g., max 48.5% improvement in NDCG@20 over LightGCN). We believe that SimpleX could not only serve as a simple strong baseline to foster future research on CF, but also shed light on the potential research direction towards improving loss function and negative sampling.

In the era of deep learning, modeling for most NLP tasks has converged to several mainstream paradigms. For example, we usually adopt the sequence labeling paradigm to solve a bundle of tasks such as POS-tagging, NER, Chunking, and adopt the classification paradigm to solve tasks like sentiment analysis. With the rapid progress of pre-trained language models, recent years have observed a rising trend of Paradigm Shift, which is solving one NLP task by reformulating it as another one. Paradigm shift has achieved great success on many tasks, becoming a promising way to improve model performance. Moreover, some of these paradigms have shown great potential to unify a large number of NLP tasks, making it possible to build a single model to handle diverse tasks. In this paper, we review such phenomenon of paradigm shifts in recent years, highlighting several paradigms that have the potential to solve different NLP tasks.

Entity linking (EL) is the process of linking entity mentions appearing in web text with their corresponding entities in a knowledge base. EL plays an important role in the fields of knowledge engineering and data mining, underlying a variety of downstream applications such as knowledge base population, content analysis, relation extraction, and question answering. In recent years, deep learning (DL), which has achieved tremendous success in various domains, has also been leveraged in EL methods to surpass traditional machine learning based methods and yield the state-of-the-art performance. In this survey, we present a comprehensive review and analysis of existing DL based EL methods. First of all, we propose a new taxonomy, which organizes existing DL based EL methods using three axes: embedding, feature, and algorithm. Then we systematically survey the representative EL methods along the three axes of the taxonomy. Later, we introduce ten commonly used EL data sets and give a quantitative performance analysis of DL based EL methods over these data sets. Finally, we discuss the remaining limitations of existing methods and highlight some promising future directions.

Abstract-- Future Moon bases will likely be constructed using resources mined from the surface of the Moon. The difficulty of maintaining a human workforce on the Moon and communications lag with Earth means that mining will need to be conducted using collaborative robots with a high degree of autonomy. In this paper, we explore the utility of robotic vision towards addressing several major challenges in autonomous mining in the lunar environment: lack of satellite positioning systems, navigation in hazardous terrain, and delicate robot interactions. The competition provided a simulated lunar environment that exhibits the complexities alluded to above. This argues for a high degree of intelligence on each agent and a robust multi-robot The need to transport resources from Earth is a serious coordination system to ensure long-term operation. In-Situ Resource some of the key challenges towards autonomous robots Utilisation (ISRU), where resources are extracted on for collaborative space mining: lack of satellite positioning other astronomical objects and exploited to support longer systems, navigation in hazardous terrain, and the need for and deeper space missions, has been proposed as a way to delicate robot interactions.

The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. A recent New York Times article concludes that new AI-powered automation tools such as Codex for software developers will not eliminate jobs but simply be a welcome aid to augment programmer productivity. This is consistent with the argument we're increasingly hearing that people and AI have different strengths and there will be appropriate roles for each. As discussed in a Harvard Business Review story: "AI-based machines are fast, more accurate, and consistently rational, but they aren't intuitive, emotional, or culturally sensitive." The belief is that "AI plus humans" is something of a centaur, greater than either one operating alone.

Let us consider the succession of neurons in Figure 1. We update the weight w (1) by calculating the derivative of Loss function respect to it. For simplicity, we assume that the bias term b is zero for each neuron. Let's ask, what can stop or slow down the gradient? We analyze the product a / z.

NVIDIA recently began inviting developers to test the newest build for DLSS (Deep Learning Super Sampling) and submit their experiences and findings to the developer forum on NVIDIA's site. NVIDIA DLSS is "a deep learning neural network that boosts frame rates and generates beautiful, sharp images for your games. It gives you the performance headroom to maximize ray tracing settings and increase output resolution. DLSS is powered by dedicated AI processors on RTX GPUs called Tensor Cores." NVIDIA is enabling developers to explore and evaluate experimental AI models for Deep Learning Super Sampling (DLSS).

An artificial intelligence program currently in development could very well be the saving grace of the human race. The results on the research of the deep-learning algorithm are being reported in a research paper, which is headed by applied mathematics professor Chris Bauch of the University of Waterloo. According to Phys.org, the paper is trying to look for specific climate crisis events called "tipping points," which are situations when humanity can no longer change the course of devastating climate change. Bauch stated that he and his team has found that their new artificial intelligence algorithm is able to predict these tipping points more accurately than before, but also offer new information at what the state of the world will be past the tipping point. A few of these "tipping points'' that Bauch's team talks about include Arctic permafrost.

Traditionally, policymakers and urban planners haven't had access to city data that can reveal complex patterns and relationships between factors that influence urban development. In some cases, data is too laborious or costly to measure at frequent time intervals, and in others, unexpected or unforeseen circumstances such as a pandemic like COVID-19 are responsible for invalidating earlier forecasts. But this is changing rapidly, with emerging technologies unlocking new possibilities for urban planning. Advances in emerging technologies like Artificial Intelligence and Machine Learning can help us understand our cities better and derive useful insights from real-time data collected through automated models that provide a much closer view of the situation on-ground compared to traditional approaches. These insights can properly assess public interests and help policymakers in making decisions that are more sustainable.