The Transformer architecture has led to significant gains in machine translation. However, most studies focus on only sentence-level translation without considering the context dependency within documents, leading to the inadequacy of document-level coherence. Some recent research tried to mitigate this issue by introducing an additional context encoder or translating with multiple sentences or even the entire document. Such methods may lose the information on the target side or have an increasing computational complexity as documents get longer. To address such problems, we introduce a recurrent memory unit to the vanilla Transformer, which supports the information exchange between the sentence and previous context. The memory unit is recurrently updated by acquiring information from sentences, and passing the aggregated knowledge back to subsequent sentence states. We follow a two-stage training strategy, in which the model is first trained at the sentence level and then finetuned for document-level translation. We conduct experiments on three popular datasets for document-level machine translation and our model has an average improvement of 0.91 s-BLEU over the sentence-level baseline. We also achieve state-of-the-art results on TED and News, outperforming the previous work by 0.36 s-BLEU and 1.49 d-BLEU on average.

The Covid-19 pandemic has led to infodemic of low quality information leading to poor health decisions. Combating the outcomes of this infodemic is not only a question of identifying false claims, but also reasoning about the decisions individuals make. In this work we propose a holistic analysis framework connecting stance and reason analysis, and fine-grained entity level moral sentiment analysis. We study how to model the dependencies between the different level of analysis and incorporate human insights into the learning process. Experiments show that our framework provides reliable predictions even in the low-supervision settings.

Code-switching is a speech phenomenon occurring when a speaker switches language during a conversation. Despite the spontaneous nature of code-switching in conversational spoken language, most existing works collect code-switching data from read speech instead of spontaneous speech. ASCEND (A Spontaneous Chinese-English Dataset) is a high-quality Mandarin Chinese-English code-switching corpus built on spontaneous multi-turn conversational dialogue sources collected in Hong Kong. We report ASCEND's design and procedure for collecting the speech data, including annotations. ASCEND consists of 10.62 hours of clean speech, collected from 23 bilingual speakers of Chinese and English. Furthermore, we conduct baseline experiments using pre-trained wav2vec 2.0 models, achieving a best performance of 22.69\% character error rate and 27.05% mixed error rate.

Bayesian optimisation (BO) has been widely used to solve problems with expensive function evaluations. In multi-objective optimisation problems, BO aims to find a set of approximated Pareto optimal solutions. There are typically two ways to build surrogates in multi-objective BO: One surrogate by aggregating objective functions (by using a scalarising function, also called mono-surrogate approach) and multiple surrogates (for each objective function, also called multi-surrogate approach). In both approaches, an acquisition function (AF) is used to guide the search process. Mono-surrogate has the advantage that only one model is used, however, the approach has two major limitations. Firstly, the fitness landscape of the scalarising function and the objective functions may not be similar. Secondly, the approach assumes that the scalarising function distribution is Gaussian, and thus a closed-form expression of the AF can be used. In this work, we overcome these limitations by building a surrogate model for each objective function and show that the scalarising function distribution is not Gaussian. We approximate the distribution using Generalised extreme value distribution. The results and comparison with existing approaches on standard benchmark and real-world optimisation problems show the potential of the multi-surrogate approach.

There used to be a time where video games were sneered at and overlooked by the culturati as lowbrow schlock but games are, and always have been, a lively and responsive form of artistic expression. It's not always immediately clear when a game was made in Australia, which makes it a little harder to celebrate homegrown hits – which we should do, because we have a thriving community of developers who punch well above their weight. The Australian independent games scene is vibrant, dynamic and overdue an apology. As I have (graciously, selflessly) decided, we're all going to yank games from the declasse and appreciate them properly – so here are 10 great Australian-made games, all variously ruminative, charming, effervescent, sincere, generous, visceral, cheeky, and beautiful. Glad we have that sorted.

One of the most promising applications of quantum computing is the processing of graphical data like images. Here, we investigate the possibility of realizing a quantum pattern recognition protocol based on swap test, and use the IBMQ noisy intermediate-scale quantum (NISQ) devices to verify the idea. We find that with a two-qubit protocol, swap test can efficiently detect the similarity between two patterns with good fidelity, though for three or more qubits the noise in the real devices becomes detrimental. To mitigate this noise effect, we resort to destructive swap test, which shows an improved performance for three-qubit states. Due to limited cloud access to larger IBMQ processors, we take a segment-wise approach to apply the destructive swap test on higher dimensional images. In this case, we define an average overlap measure which shows faithfulness to distinguish between two very different or very similar patterns when simulated on real IBMQ processors. As test images, we use binary images with simple patterns, greyscale MNIST numbers and MNIST fashion images, as well as binary images of human blood vessel obtained from magnetic resonance imaging (MRI). We also present an experimental set up for applying destructive swap test using the nitrogen vacancy centre (NVs) in diamond. Our experimental data show high fidelity for single qubit states. Lastly, we propose a protocol inspired from quantum associative memory, which works in an analogous way to supervised learning for performing quantum pattern recognition using destructive swap test.

Medical coding (MC) is an essential pre-requisite for reliable data retrieval and reporting. Given a free-text reported term (RT) such as "pain of right thigh to the knee", the task is to identify the matching lowest-level term (LLT) - in this case "unilateral leg pain" - from a very large and continuously growing repository of standardized medical terms. However, automating this task is challenging due to a large number of LLT codes (as of writing over 80,000), limited availability of training data for long tail/emerging classes, and the general high accuracy demands of the medical domain. With this paper, we introduce the MC task, discuss its challenges, and present a novel approach called xTARS that combines traditional BERT-based classification with a recent zero/few-shot learning approach (TARS). We present extensive experiments that show that our combined approach outperforms strong baselines, especially in the few-shot regime. The approach is developed and deployed at Bayer, live since November 2021. As we believe our approach potentially promising beyond MC, and to ensure reproducibility, we release the code to the research community.

Accurate depth-sensing plays a crucial role in securing a high success rate of robotic harvesting in natural orchard environments. Solid-state LiDAR (SSL), a recently introduced LiDAR technique, can perceive high-resolution geometric information of the scenes, which can be potential utilised to receive accurate depth information. Meanwhile, the fusion of the sensory information from LiDAR and camera can significantly enhance the sensing ability of the harvesting robots. This work introduces a LiDAR-camera fusion-based visual sensing and perception strategy to perform accurate fruit localisation for a harvesting robot in the apple orchards. Two SOTA extrinsic calibration methods, target-based and targetless-based, are applied and evaluated to obtain the accurate extrinsic matrix between the LiDAR and camera. With the extrinsic calibration, the point clouds and color images are fused to perform fruit localisation using a one-stage instance segmentation network. Experimental shows that LiDAR-camera achieves better quality on visual sensing in the natural environments. Meanwhile, introducing the LiDAR-camera fusion largely improves the accuracy and robustness of the fruit localisation. Specifically, the standard deviations of fruit localisation by using LiDAR-camera at 0.5 m, 1.2 m, and 1.8 m are 0.245, 0.227, and 0.275 cm respectively. These measurement error is only one one fifth of that from Realsense D455. Lastly, we have attached our visualised point cloud to demonstrate the highly accurate sensing method.

Deep Learning Research Scientist at Engineroom Remote › Worldwide, 100% remote position (Posted Apr 30 2022) Job description You will work in our research team to extend the capabilities of neural networks applied to video / audio / text understanding and common sense reasoning. You will perform research on recurrent networks, attention policies, multi-task learning and scaling up training. You will have expert understanding of NIPS, ICLR, CVPR, PAMI. You will help define and shape a platform with practical real world client applications from Day 1 at the forefront of Deep Learning and AI. We have used our expertise to: Increase crop yields, predict credit defaults, identify dark matter, render blockbuster VFX, develop & simulate medical breakthroughs, calculate re-entry profiles for spacecraft, give machine vision to Robots and UAV's, and much more......... WHAT WE OFFER: Full time, Part-time, Contract Roles.

Marley Spoon is bringing delightful, market fresh and easy cooking back to the people, while re-inventing the global food supply chain to reduce unethical food waste. We are connecting consumers with quality producers, providing fresh ingredients directly to customer's homes and we are present in Europe, the US and Australia, shipping more than 140 000 food boxes every week. We are a lean team working closely with the Data Warehousing, Data Science and DevOps Teams. The main focus of this team is to build and provide Services and Infrastructure to the Data Science Team to move models into production while following best practices. As a data driven company this position is a vital one for the growth and success of Marley Spoon.