This paper considers the problem of designating navigation goal locations for interactive mobile robots. We investigate a point-andclick interface, implemented with an Augmented Reality (AR) headset. The cameras on the AR headset are used to detect natural pointing gestures performed by the user. The selected goal is visualized through the AR headset, allowing the users to adjust the goal location if desired. We conduct a user study in which participants set consecutive navigation goals for the robot using three different interfaces: AR Point&Click, Person Following and Tablet (birdeye map view). Results show that the proposed AR Point&Click interface improved the perceived accuracy, efficiency and reduced mental load compared to the baseline tablet interface, and it performed on-par to the Person Following method. These results show that the AR Point&Click is a feasible interaction model for setting navigation goals.

Response diversity has become an important criterion for evaluating the quality of open-domain dialogue generation models. However, current evaluation metrics for response diversity often fail to capture the semantic diversity of generated responses, as they mainly consider lexical aspects of the generated responses. In this paper, we introduce a new automatic evaluation metric to measure the semantic diversity of generated responses. Through human evaluation, we demonstrate that our proposed metric captures human judgments on response diversity better than existing lexical-level diversity metrics. Furthermore, motivated by analyzing an existing dialogue dataset, we propose a simple yet effective learning method that improves the semantic diversity of generated responses. Our learning method weights training samples based on the semantic distribution of the training set. We show that our learning method improves response diversity and coherency better than other baseline methods through automatic and human evaluation.

Bayesian optimisation (BO) algorithms have shown remarkable success in applications involving expensive black-box functions. Traditionally BO has been set as a sequential decision-making process which estimates the utility of query points via an acquisition function and a prior over functions, such as a Gaussian process. Recently, however, a reformulation of BO via density-ratio estimation (BORE) allowed reinterpreting the acquisition function as a probabilistic binary classifier, removing the need for an explicit prior over functions and increasing scalability. In this paper, we present a theoretical analysis of BORE's regret and an extension of the algorithm with improved uncertainty estimates. We also show that BORE can be naturally extended to a batch optimisation setting by recasting the problem as approximate Bayesian inference. The resulting algorithms come equipped with theoretical performance guarantees and are assessed against other batch and sequential BO baselines in a series of experiments.

Programming education in India faces an uphill task of educating two-plus million students who enroll in degree programs with coding as a core skill.10 Fulfilling this demand presents unique challenges due to inadequate infrastructure and the unavailability of technical content in regional languages6 with unfortunate outcomes: more than 90% of Indian graduates have coding skills inadequate for IT roles, and more than 37% struggle to write code that even compiles.13 Studies indicate a steep decline in coding skills between graduates from top-100 colleges and the rest.12 While concerning, since only a tiny minority of students enroll in top-tier colleges, this is not entirely surprising. A recent study indicates that even experienced instructors at non-top-tier colleges in India struggle to write code.10

The US Court of Appeals has upheld previous rulings that AIs cannot hold patents for inventions. AIs are increasingly being used to make new discoveries but, under most patent laws, a human must be listed as the patent holder for inventions. Dr Stephen Thaler created a device called DABUS that consists of neural networks and has been used to invent an emergency warning light, a food container that improves grip and heat transfer, and more. Thaler believes that AIs should be patent holders and has launched numerous cases in at least 15 countries to argue the case. So far, the cases in the UK, US, and New Zealand have all been rejected.

Established in 2010, Knorex is a cutting-edge advertising technology MNC with offices across USA, Australia, China, Singapore, Vietnam, India, Thailand and Malaysia. Knorex provides Precision Performance Marketing products and solutions to the world's leading brands and media agencies. With its full-stack platform, Knorex XPO (Unify to Simplify - Knorex XPO - The Ultimate Universal Marketing Platform to Power Business Growth Knorex.com)

Since the first AI-HRI held at the 2014 AAAI Fall Symposium Series, a lot of the presented research and discussions have emphasized how artificial intelligence (AI) developments can benefit human-robot interaction (HRI). This portrays HRI as an application, a source of domain-specific problems to solve, to the AI community. Likewise, this portrays AI as a tool, a source of solutions available for relevant problems, to the HRI community. However, members of the AI-HRI research community will point out that the relationship has a deeper synergy than matchmaking problems and solutions -- there are insights from each field that impact how the other one thinks about the world and performs scientific research. There is no greater opportunity for sharing perspectives at the moment than human-aware AI, which studies how to account for the fact that people are more than a source of data or part of an algorithm. We will explore how AI-HRI can change the way researchers think about human-aware AI, from observation through validation, to make even the algorithmic design process human-aware.

However, working ontologists are often surprised by how challenging and slow it can be to develop ontologies. Here, with a particular emphasis on the sorts of ontologies that are content-heavy and intended to be shared across a community of users (reference ontologies), we propose that a significant and heretofore under-emphasised contributor of challenges during ontology development is the need to create, or bring about, consensus in the face of disagreement. For this reason reference ontology development cannot be automated, at least within the limitations of existing AI approaches. Further, for the same reason ontologists are required to have specific social-negotiating skills which are currently lacking in most technical curricula.

Sanskrit Word Segmentation (SWS) is essential in making digitized texts available and in deploying downstream tasks. It is, however, non-trivial because of the sandhi phenomenon that modifies the characters at the word boundaries, and needs special treatment. Existing lexicon driven approaches for SWS make use of Sanskrit Heritage Reader, a lexicon-driven shallow parser, to generate the complete candidate solution space, over which various methods are applied to produce the most valid solution. However, these approaches fail while encountering out-of-vocabulary tokens. On the other hand, purely engineering methods for SWS have made use of recent advances in deep learning, but cannot make use of the latent word information on availability. To mitigate the shortcomings of both families of approaches, we propose Transformer based Linguistically Informed Sanskrit Tokenizer (TransLIST) consisting of (1) a module that encodes the character input along with latent-word information, which takes into account the sandhi phenomenon specific to SWS and is apt to work with partial or no candidate solutions, (2) a novel soft-masked attention to prioritize potential candidate words and (3) a novel path ranking algorithm to rectify the corrupted predictions. Experiments on the benchmark datasets for SWS show that TransLIST outperforms the current state-of-the-art system by an average 7.2 points absolute gain in terms of perfect match (PM) metric. The codebase and datasets are publicly available at https://github.com/rsingha108/TransLIST

The Transformer architecture aggregates input information through the self-attention mechanism, but there is no clear understanding of how this information is mixed across the entire model. Additionally, recent works have demonstrated that attention weights alone are not enough to describe the flow of information. In this paper, we consider the whole attention block -- multi-head attention, residual connection, and layer normalization -- and define a metric to measure token-to-token interactions within each layer. Then, we aggregate layer-wise interpretations to provide input attribution scores for model predictions. Experimentally, we show that our method, ALTI (Aggregation of Layer-wise Token-to-token Interactions), provides more faithful explanations and increased robustness than gradient-based methods.