Successful Artificial Intelligence systems often require numerous labeled data to extract information from document images. In this paper, we investigate the problem of improving the performance of Artificial Intelligence systems in understanding document images, especially in cases where training data is limited. We address the problem by proposing a novel finetuning method using reinforcement learning. Our approach treats the Information Extraction model as a policy network and uses policy gradient training to update the model to maximize combined reward functions that complement the traditional cross-entropy losses. Our experiments on four datasets using labels and expert feedback demonstrate that our finetuning mechanism consistently improves the performance of a state-of-the-art information extractor, especially in the small training data regime.

The variational quantum eigensolver (VQE) is a leading strategy that exploits noisy intermediate-scale quantum (NISQ) machines to tackle chemical problems outperforming classical approaches. To gain such computational advantages on large-scale problems, a feasible solution is the QUantum DIstributed Optimization (QUDIO) scheme, which partitions the original problem into $K$ subproblems and allocates them to $K$ quantum machines followed by the parallel optimization. Despite the provable acceleration ratio, the efficiency of QUDIO may heavily degrade by the synchronization operation. To conquer this issue, here we propose Shuffle-QUDIO to involve shuffle operations into local Hamiltonians during the quantum distributed optimization. Compared with QUDIO, Shuffle-QUDIO significantly reduces the communication frequency among quantum processors and simultaneously achieves better trainability. Particularly, we prove that Shuffle-QUDIO enables a faster convergence rate over QUDIO. Extensive numerical experiments are conducted to verify that Shuffle-QUDIO allows both a wall-clock time speedup and low approximation error in the tasks of estimating the ground state energy of molecule. We empirically demonstrate that our proposal can be seamlessly integrated with other acceleration techniques, such as operator grouping, to further improve the efficacy of VQE.

The task of empathetic response generation aims to understand what feelings a speaker expresses on his/her experiences and then reply to the speaker appropriately. To solve the task, it is essential to model the content-emotion duality of a dialogue, which is composed of the content view (i.e., what personal experiences are described) and the emotion view (i.e., the feelings of the speaker on these experiences). To this end, we design a framework to model the Content-Emotion Duality (CEDual) via disentanglement for empathetic response generation. With disentanglement, we encode the dialogue history from both the content and emotion views, and then generate the empathetic response based on the disentangled representations, thereby both the content and emotion information of the dialogue history can be embedded in the generated response. The experiments on the benchmark dataset EMPATHETICDIALOGUES show that the CEDual model achieves state-of-the-art performance on both automatic and human metrics, and it also generates more empathetic responses than previous methods.

Text generation is of great importance to many natural language processing applications. However, maximization-based decoding methods (e.g., beam search) of neural language models often lead to degenerate solutions--the generated text is unnatural and contains undesirable repetitions. Existing approaches introduce stochasticity via sampling or modify training objectives to decrease the probabilities of certain tokens (e.g., unlikelihood training). However, they often lead to solutions that lack coherence. In this work, we show that an underlying reason for model degeneration is the anisotropic distribution of token representations. We present a contrastive solution: (i) SimCTG, a contrastive training objective to calibrate the model's representation space, and (ii) a decoding method--contrastive search--to encourage diversity while maintaining coherence in the generated text. Extensive experiments and analyses on three benchmarks from two languages demonstrate that our proposed approach significantly outperforms current state-of-the-art text generation methods as evaluated by both human and automatic metrics.

In traditional astronomies across the world, groups of stars in the night sky were linked into constellations -- symbolic representations rich in meaning and with practical roles. In some sky cultures, constellations are represented as line (or connect-the-dot) figures, which are spatial networks drawn over the fixed background of stars. We analyse 1802 line figures from 56 sky cultures spanning all continents, in terms of their network, spatial, and brightness features, and ask what associations exist between these visual features and culture type or sky region. First, an embedded map of constellations is learnt, to show clusters of line figures. We then form the network of constellations (as linked by their similarity), to study how similar cultures are by computing their assortativity (or homophily) over the network. Finally, we measure the diversity (or entropy) index for the set of constellations drawn per sky region. Our results show distinct types of line figures, and that many folk astronomies with oral traditions have widespread similarities in constellation design, which do not align with cultural ancestry. In a minority of sky regions, certain line designs appear universal, but this is not the norm: in the majority of sky regions, the line geometries are diverse.

This study develops a framework for unmanned aerial systems (UASs) to monitor fall hazard prevention systems near unprotected edges and openings in high-rise building projects. A three-step machine-learning-based framework was developed and tested to detect guardrail posts from the images captured by UAS. First, a guardrail detector was trained to localize the candidate locations of posts supporting the guardrail. Since images were used in this process collected from an actual job site, several false detections were identified. Therefore, additional constraints were introduced in the following steps to filter out false detections. Second, the research team applied a horizontal line detector to the image to properly detect floors and remove the detections that were not close to the floors. Finally, since the guardrail posts are installed with approximately normal distribution between each post, the space between them was estimated and used to find the most likely distance between the two posts. The research team used various combinations of the developed approaches to monitor guardrail systems in the captured images from a high-rise building project. Comparing the precision and recall metrics indicated that the cascade classifier achieves better performance with floor detection and guardrail spacing estimation. The research outcomes illustrate that the proposed guardrail recognition system can improve the assessment of guardrails and facilitate the safety engineer's task of identifying fall hazards in high-rise building projects.

Due to the rapid development of non-face-to-face services due to the corona virus, commerce through the Internet, such as sales and reservations, is increasing very rapidly. Consumers also post reviews, suggestions, or judgments about goods or services on the website. The review data directly used by consumers provides positive feedback and nice impact to consumers, such as creating business value. Therefore, analysing review data is very important from a marketing point of view. Our research suggests a new way to find factors for customer satisfaction through review data. We applied a method to find factors for customer satisfaction by mixing and using the data mining technique, which is a big data analysis method, and the natural language processing technique, which is a language processing method, in our research. Unlike many studies on customer satisfaction that have been conducted in the past, our research has a novelty of the thesis by using various techniques. And as a result of the analysis, the results of our experiments were very accurate.

Methods have recently been proposed that densely segment 3D volumes into classes using only color images and expert supervision in the form of sparse semantically annotated pixels. While impressive, these methods still require a relatively large amount of supervision and segmenting an object can take several minutes in practice. Such systems typically only optimize their representation on the particular scene they are fitting, without leveraging any prior information from previously seen images. In this paper, we propose to use features extracted with models trained on large existing datasets to improve segmentation performance. We bake this feature representation into a Neural Radiance Field (NeRF) by volumetrically rendering feature maps and supervising on features extracted from each input image. We show that by baking this representation into the NeRF, we make the subsequent classification task much easier. Our experiments show that our method achieves higher segmentation accuracy with fewer semantic annotations than existing methods over a wide range of scenes.

There has been much hand-wringing about the crisis of the humanities, and recent breakthroughs in artificial intelligence have added to the angst. It is not only truck drivers whose jobs are threatened by automation. Deep-learning algorithms are also entering the domain of creative work. And now, they are demonstrating proficiency in the tasks that occupy humanities professors when they are not giving lectures: namely, writing papers and submitting them for publication in academic journals. Could academic publishing be automated?

The fourth industrial revolution is underway and it's threatening to wipe out nearly half the jobs in Australia. This latest round is characterised by intelligent robots and machine learning and PricewaterhouseCoopers economist Jeremy Thorpe said it's going to completely reshape the Australian jobs market. "Over the next 20 years approximately 44 per cent of Australia's jobs, that's more than 5 million jobs, are at risk of being disrupted by technology, whether that's digitisation or automation," he said. It seems only a matter of time before intelligence breaks free of its biological bonds - Joe Gelonesi talks to two philosophers with profoundly different visions of technology and the future. Stefan Hajkowicz, who is the principal scientist at the CSIRO, says it's white collar workers who are about to feel the pain.