The spread of the Sars-COV-2 pandemic in 2020 had a huge impact on the life course of all of us. This rapid spread has also caused an increase in the research production in topics related to COVID-19 with regard to different aspects. Italy has, unfortunately, been one of the first countries to be massively involved in the outbreak of the disease. In this paper we present an extensive scientometric analysis of the research production both at global (entire literature produced in the first 2 years after the beginning of the pandemic) and local level (COVID-19 literature produced by authors with an Italian affiliation). Our results showed that US and China are the most active countries in terms of number of publications and that the number of collaborations between institutions varies according to geographical distance. Moreover, we identified the medical-biological as the fields with the greatest growth in terms of literature production. Furthermore, we also better explored the relationship between the number of citations and variables obtained from the data set (e.g. number of authors per article). Using multiple correspondence analysis and quantile regression we shed light on the role of journal topics and impact factor, the type of article, the field of study and how these elements affect citations.

The technical progression of artificial intelligence (AI) research has been built on breakthroughs in fields such as computer science, statistics, and mathematics. However, in the past decade AI researchers have increasingly looked to the social sciences, turning to human interactions to solve the challenges of model development. Paying crowdsourcing workers to generate or curate data, or data enrichment, has become indispensable for many areas of AI research, from natural language processing to reinforcement learning from human feedback (RLHF). Other fields that routinely interact with crowdsourcing workers, such as Psychology, have developed common governance requirements and norms to ensure research is undertaken ethically. This study explores how, and to what extent, comparable research ethics requirements and norms have developed for AI research and data enrichment. We focus on the approach taken by two leading conferences: ICLR and NeurIPS, and journal publisher Springer. In a longitudinal study of accepted papers, and via a comparison with Psychology and CHI papers, this work finds that leading AI venues have begun to establish protocols for human data collection, but these are are inconsistently followed by authors. Whilst Psychology papers engaging with crowdsourcing workers frequently disclose ethics reviews, payment data, demographic data and other information, similar disclosures are far less common in leading AI venues despite similar guidance. The work concludes with hypotheses to explain these gaps in research ethics practices and considerations for its implications.

We study the design of loss functions for click-through rates (CTR) to optimize (social) welfare in advertising auctions. Existing works either only focus on CTR predictions without consideration of business objectives (e.g., welfare) in auctions or assume that the distribution over the participants' expected cost-per-impression (eCPM) is known a priori, then use various additional assumptions on the parametric form of the distribution to derive loss functions for predicting CTRs. In this work, we bring back the welfare objectives of ad auctions into CTR predictions and propose a novel weighted rankloss to train the CTR model. Compared to existing literature, our approach provides a provable guarantee on welfare but without assumptions on the eCPMs' distribution while also avoiding the intractability of naively applying existing learning-to-rank methods. Further, we propose a theoretically justifiable technique for calibrating the losses using labels generated from a teacher network, only assuming that the teacher network has bounded $\ell_2$ generalization error. Finally, we demonstrate the advantages of the proposed loss on synthetic and real-world data.

Low speed autonomous shuttles emulating SAE Level L4 automated driving using human driver assisted autonomy have been operating in geo-fenced areas in several cities in the US and the rest of the world. These autonomous vehicles (AV) are operated by small to mid-sized technology companies that do not have the resources of automotive OEMs for carrying out exhaustive, comprehensive testing of their AV technology solutions before public road deployment. Due to the low speed of operation and hence not operating on roads containing highways, the base vehicles of these AV shuttles are not required to go through rigorous certification tests. The way the driver assisted AV technology is tested and allowed for public road deployment is continuously evolving but is not standardized and shows differences between the different states where these vehicles operate. Currently, AVs and AV shuttles deployed on public roads are using these deployments for testing and improving their technology. However, this is not the right approach. Safe and extensive testing in a lab and controlled test environment including Model-in-the-Loop (MiL), Hardware-in-the-Loop (HiL) and Autonomous-Vehicle-in-the-Loop (AViL) testing should be the prerequisite to such public road deployments. This paper presents three dimensional virtual modeling of an AV shuttle deployment site and simulation testing in this virtual environment. We have two deployment sites in Columbus of these AV shuttles through the Department of Transportation funded Smart City Challenge project named Smart Columbus. The Linden residential area AV shuttle deployment site of Smart Columbus is used as the specific example for illustrating the AV testing method proposed in this paper.

Low probability of detection (LPD) has recently emerged as a means to enhance the privacy and security of wireless networks. Unlike existing wireless security techniques, LPD measures aim to conceal the entire existence of wireless communication instead of safeguarding the information transmitted from users. Motivated by LPD communication, in this paper, we study a privacy-preserving and distributed framework based on graph neural networks to minimise the detectability of a wireless ad-hoc network as a whole and predict an optimal communication region for each node in the wireless network, allowing them to communicate while remaining undetected from external actors. We also demonstrate the effectiveness of the proposed method in terms of two performance measures, i.e., mean absolute error and median absolute error.

Theory of Mind (ToM)$\unicode{x2014}$the ability to reason about the mental states of other people$\unicode{x2014}$is a key element of our social intelligence. Yet, despite their ever more impressive performance, large-scale neural language models still lack basic theory of mind capabilities out-of-the-box. We posit that simply scaling up models will not imbue them with theory of mind due to the inherently symbolic and implicit nature of the phenomenon, and instead investigate an alternative: can we design a decoding-time algorithm that enhances theory of mind of off-the-shelf neural language models without explicit supervision? We present SymbolicToM, a plug-and-play approach to reason about the belief states of multiple characters in reading comprehension tasks via explicit symbolic representation. More concretely, our approach tracks each entity's beliefs, their estimation of other entities' beliefs, and higher-order levels of reasoning, all through graphical representations, allowing for more precise and interpretable reasoning than previous approaches. Empirical results on the well-known ToMi benchmark (Le et al., 2019) demonstrate that SymbolicToM dramatically enhances off-the-shelf neural networks' theory of mind in a zero-shot setting while showing robust out-of-distribution performance compared to supervised baselines. Our work also reveals spurious patterns in existing theory of mind benchmarks, emphasizing the importance of out-of-distribution evaluation and methods that do not overfit a particular dataset.

Reinforcement learning based dialogue policies are typically trained in interaction with a user simulator. To obtain an effective and robust policy, this simulator should generate user behaviour that is both realistic and varied. Current data-driven simulators are trained to accurately model the user behaviour in a dialogue corpus. We propose an alternative method using adversarial learning, with the aim to simulate realistic user behaviour with more variation. We train and evaluate several simulators on a corpus of restaurant search dialogues, and then use them to train dialogue system policies. In policy cross-evaluation experiments we demonstrate that an adversarially trained simulator produces policies with 8.3% higher success rate than those trained with a maximum likelihood simulator. Subjective results from a crowd-sourced dialogue system user evaluation confirm the effectiveness of adversarially training user simulators.

Despite the ubiquity of ambiguity in natural language, it is often ignored or deliberately removed in semantic parsing tasks, which generally assume that a given surface form has only one correct logical form. We attempt to address this shortcoming by introducing AmP, a framework, dataset, and challenge for parsing with linguistic ambiguity. We define templates and generate data for five well-documented linguistic ambiguities. Using AmP, we investigate how several few-shot semantic parsing systems handle ambiguity, introducing three new metrics. We find that large pre-trained models perform poorly at capturing the distribution of possible meanings without deliberate instruction. However, models are able to capture distribution well when ambiguity is attested in their inputs. These results motivate a call for ambiguity to be explicitly included in semantic parsing, and promotes considering the distribution of possible outputs when evaluating semantic parsing systems.

Learning the complex structure of multivariate time series has been one of the major interests across many application domains, including economics, transportation, manufacturing [Fortuin et al., 2020, Wu et al., 2021, Li et al., 2019, Zhou et al., 2021]. While there has been much progress in the data-driven learning and processing complex time series, it still remains as a challenging topic, in particular, when the data is corrupted [Cao et al., 2018, Kreindler and Lumsden, 2006, Yoon et al., 2018, Du et al., 2022]. In this paper, we consider the forecasting task which aims to make prediction of future values using historical data that may contain missing values. In addition, for many industrial problems, the time series features can be in two categories: auxiliary features (X) that provide information about the state of a system and target variables (Y) that depends on the auxiliary features and may convey valuable information. For example, in the operation of a chemical reactor, the auxiliary features include temperature, pressure and concentration of chemicals observed through a sensor network, while the target variable may include the quality of the material and throughput. We are interested in the time series problem where the data set consists of X and Y. In general, X is more readily available, as it is obtained from a sensor network, while Y may be temporally sparse since it may be expensive or difficult to collect the data. This so-called soft sensor problem has been of interest in many industrial applications [Shardt et al., 2015, Yuan et al., 2021].

In the context of structure-to-structure transformation tasks, learning sequences of discrete symbolic operations poses significant challenges due to their non-differentiability. To facilitate the learning of these symbolic sequences, we introduce a differentiable tree interpreter that compiles high-level symbolic tree operations into subsymbolic matrix operations on tensors. We present a novel Differentiable Tree Machine (DTM) architecture that integrates our interpreter with an external memory and an agent that learns to sequentially select tree operations to execute the target transformation in an end-to-end manner. With respect to out-of-distribution compositional generalization on synthetic semantic parsing and language generation tasks, DTM achieves 100% while existing baselines such as Transformer, Tree Transformer, LSTM, and Tree2Tree LSTM achieve less than 30%. DTM remains highly interpretable in addition to its perfect performance.