US Defense Secretary Lloyd Austin has issued a directive ordering the United States military to do more to protect civilians from harm in drone attacks and other combat operations. In a two-page memo to top Pentagon civilian and military officials, Austin on Thursday ordered a comprehensive overhaul of the US Defense Department's posture towards protecting civilians in conflict zones. "The protection of innocent civilians in the conduct of our operations remains vital to the ultimate success of our operations and as a significant strategic and moral imperative," the memo reads. The defence secretary asked for an action plan from the Joint Chiefs of Staff to prevent harm to civilians and improve US responses when such incidents occur. That plan is due within 90 days.

Note that the Buckeye Flats location (a) contains greater acoustic activity, a result of the nearby rapid flowing stream that produced considerable geophonic sounds. The inset (b) graphs the same data but with Buckeye Flats removed. These values (b) reflect mostly biophony. Sycamore Creek contained the greatest acoustic activity of these three. The fall contains the greatest activity although there was no consistent pattern across sites. Photos of each landscape are provided in (c).

Few-shot Learning (FSL) is aimed to make predictions based on a limited number of samples. Structured data such as knowledge graphs and ontology libraries has been leveraged to benefit the few-shot setting in various tasks. However, the priors adopted by the existing methods suffer from challenging knowledge missing, knowledge noise, and knowledge heterogeneity, which hinder the performance for few-shot learning. In this study, we explore knowledge injection for FSL with pre-trained language models and propose ontology-enhanced prompt-tuning (OntoPrompt). Specifically, we develop the ontology transformation based on the external knowledge graph to address the knowledge missing issue, which fulfills and converts structure knowledge to text. We further introduce span-sensitive knowledge injection via a visible matrix to select informative knowledge to handle the knowledge noise issue. To bridge the gap between knowledge and text, we propose a collective training algorithm to optimize representations jointly. We evaluate our proposed OntoPrompt in three tasks, including relation extraction, event extraction, and knowledge graph completion, with eight datasets. Experimental results demonstrate that our approach can obtain better few-shot performance than baselines.

In the pursuit of explaining implicit regularization in deep learning, prominent focus was given to matrix and tensor factorizations, which correspond to simplified neural networks. It was shown that these models exhibit implicit regularization towards low matrix and tensor ranks, respectively. Drawing closer to practical deep learning, the current paper theoretically analyzes the implicit regularization in hierarchical tensor factorization, a model equivalent to certain deep convolutional neural networks. Through a dynamical systems lens, we overcome challenges associated with hierarchy, and establish implicit regularization towards low hierarchical tensor rank. This translates to an implicit regularization towards locality for the associated convolutional networks. Inspired by our theory, we design explicit regularization discouraging locality, and demonstrate its ability to improve performance of modern convolutional networks on non-local tasks, in defiance of conventional wisdom by which architectural changes are needed. Our work highlights the potential of enhancing neural networks via theoretical analysis of their implicit regularization.

Recent approaches of computer vision utilize deep learning methods as they perform quite well if training and testing domains follow the same underlying data distribution. However, it has been shown that minor variations in the images that occur when using these methods in the real world can lead to unpredictable errors. Transfer learning is the area of machine learning that tries to prevent these errors. Especially, approaches that augment image data using auxiliary knowledge encoded in language embeddings or knowledge graphs (KGs) have achieved promising results in recent years. This survey focuses on visual transfer learning approaches using KGs. KGs can represent auxiliary knowledge either in an underlying graph-structured schema or in a vector-based knowledge graph embedding. Intending to enable the reader to solve visual transfer learning problems with the help of specific KG-DL configurations we start with a description of relevant modeling structures of a KG of various expressions, such as directed labeled graphs, hypergraphs, and hyper-relational graphs. We explain the notion of feature extractor, while specifically referring to visual and semantic features. We provide a broad overview of knowledge graph embedding methods and describe several joint training objectives suitable to combine them with high dimensional visual embeddings. The main section introduces four different categories on how a KG can be combined with a DL pipeline: 1) Knowledge Graph as a Reviewer; 2) Knowledge Graph as a Trainee; 3) Knowledge Graph as a Trainer; and 4) Knowledge Graph as a Peer. To help researchers find evaluation benchmarks, we provide an overview of generic KGs and a set of image processing datasets and benchmarks including various types of auxiliary knowledge. Last, we summarize related surveys and give an outlook about challenges and open issues for future research.

A computer scientist who said she was pushed out of her job at Google in December 2020 has marked the one-year anniversary of her ouster with a new research institute aiming to support the creation of ethical artificial intelligence. Timnit Gebru, a known advocate for diversity in AI, announced the launch of the Distributed Artificial Intelligence Research Institute, or DAIR. Its website describes it as "a space for independent, community-rooted AI research free from Big Tech's pervasive influence." Part of how Gebru imagines creating such research is by moving away from the Silicon Valley ethos of "move fast and break things" -- which was Facebook's internal motto, coined by Mark Zuckerberg, until 2014 -- to instead take a more deliberate approach to creating new technologies that serve marginalized communities. That includes recognizing and mitigating technologies' potentials for harm from the beginning of their creation process, rather than after they've already caused damage to those communities, Gebru told NBC News.

Welcome to the first episode of New voices in AI! You can find David on Twitter @davlanade and find out more about Masakhane here. The music used is'Wholesome' by Kevin MacLeod, Licensed under Creative Commons Daly: Hello and welcome to new voices in AI, this a new series from AIhub where we celebrate the voices PhD students, early career researchers, and those with a new perspective on AI. And without further ado, let's begin. First up, a big welcome to our very first guest on "New voices in AI" and if you could introduce yourself, who are you? Adelani: Thank you very much for having me. So, Masakhane is this grassroots organization, whose mission is to strengthen and spur NLP research in African languages, by Africans for Africans, so, and currently the organization we are majorly operating on Slack we already have over 1000 Members. Of course, not everyone is active but we have more than 100 or close to 100 active members as well, yeah. So how did, how did you get into AI?

Apart from its distinct epidemiology, the natural behavior, treatment, and prognosis are different from other head and neck cancers. With the growing trend of artificial intelligence (AI), especially deep learning (DL), in head and neck cancer care, we sought to explore the unique clinical application and implementation direction of AI in the management of NPC. Methods: The search protocol was performed to collect publications using AI, machine learning (ML) and DL in NPC management from PubMed, Scopus and Embase. The articles were filtered using inclusion and exclusion criteria, and the quality of the papers was assessed. Data were extracted from the finalized articles.

Analysis For campaigners hoping that 2022 could be the year that deepfaked imagery falls within a stricter legal purview, the early indicators are unpromising. Last Thursday the European Parliament ratified amendments to the Digital Services Act (DSA, due to take effect in 2023), in regards to the dissemination of deepfakes. The modifications address deepfakes across two sections, each directly related to online advertising: amendment 1709 pertaining to Article 30, and a related amendment to article 63. 'Where a very large online platform becomes aware that a piece of content is a generated or manipulated image, audio or video content that appreciably resembles existing persons, objects, places or other entities or events and falsely appears to a person to be authentic or truthful (deep fakes), the provider shall label the content in a way that informs that the content is inauthentic and that is clearly visible for the recipient of the services.' The second adds text to the existing article 63, which is itself mainly concerned with increasing the transparency of large advertising platforms. 'In addition, very large online platforms should label any known deep fake videos, audio or other files.'

Microsoft continues to make significant investments in deep learning, computer vision, and AI. The Microsoft Maps Team has been leveraging that investment to identify map features at scale and produce high-quality building footprint data sets with the overall goal to add to the OpenStreetMap and MissingMaps humanitarian efforts. As of this post, the following locations are available and Microsoft offers access to this data under the Open Data Commons Open Database License (ODbL). Country/Region Million buildings United States of America 129.6 Nigeria and Kenya 50.5 South America 44.5 Uganda and Tanzania 17.9 Canada 11.8 Australia 11.3 As you might expect, the vintage of the footprints depends on the collection date of the underlying imagery. Bing Maps Imagery is a composite of multiple sources with different capture dates (ranging 2012 to 2021).