This paper proposes a tool for efficiently constructing high-quality parallel corpora with minimizing human labor and making this tool publicly available. Our proposed construction process is based on neural machine translation (NMT) to allow for it to not only coexist with human translation, but also improve its efficiency by combining data quality control with human translation in a data-centric approach.

Facebook trained its artificial intelligence systems to detect and block any future attempt to livestream a shooting spree with "police/military body cams footage," and other violent material, in the aftermath of the Christchurch terror attack. The emergency exercise – detailed in corporate papers leaked by whistleblower Frances Haugen – followed the March 2019 mass murder in the New Zealand city, described internally as "a watershed moment" for the Facebook Live video service. The white supremacist attacker was able to broadcast a 17-minute live stream of the attack on two mosques that was not detected by the company's systems, allowing it to be swiftly replicated online. "It was clear that Live was a vulnerable surface which can be repurposed by bad actors to cause societal harm," the leaked review stated. "Since this event, we've faced international media pressure and have seen regulatory and legal risks increase on Facebook increase considerably."

Atlassian can hire people in any country where we have a legal entity, assuming candidates have eligible working rights and a sufficient timezone overlap with their team. As our offices re-open, Atlassians can choose to work remotely or return to an office, unless it's necessary for the role to be performed in the office. Interviews and onboarding are conducted virtually, a part of being a distributed-first company. With a sufficient timezone overlap with the team, we're able to hire eligible candidates for this role from any location in Australia and New Zealand. If this sparks your interest, apply today and chat with our friendly Recruitment team further.

The U.S. Centers for Disease Control and Prevention estimates that one in seven children in the United States experienced abuse or neglect in the past year. Child protective services agencies around the nation receive a high number of reports each year (about 4.4 million in 2019) of alleged neglect or abuse. With so many cases, some agencies are implementing machine learning models to help child welfare specialists screen cases and determine which to recommend for further investigation. But these models don't do any good if the humans they are intended to help don't understand or trust their outputs. Researchers at MIT and elsewhere launched a research project to identify and tackle machine learning usability challenges in child welfare screening.

Despite strong performance in numerous applications, the fragility of deep learning to input perturbations has raised serious questions about its use in safety-critical domains. While adversarial training can mitigate this issue in practice, state-of-the-art methods are increasingly application-dependent, heuristic in nature, and suffer from fundamental trade-offs between nominal performance and robustness. Moreover, the problem of finding worst-case perturbations is non-convex and underparameterized, both of which engender a non-favorable optimization landscape. Thus, there is a gap between the theory and practice of adversarial training, particularly with respect to when and why adversarial training works. In this paper, we take a constrained learning approach to address these questions and to provide a theoretical foundation for robust learning. In particular, we leverage semi-infinite optimization and non-convex duality theory to show that adversarial training is equivalent to a statistical problem over perturbation distributions, which we characterize completely. Notably, we show that a myriad of previous robust training techniques can be recovered for particular, sub-optimal choices of these distributions. Using these insights, we then propose a hybrid Langevin Monte Carlo approach of which several common algorithms (e.g., PGD) are special cases. Finally, we show that our approach can mitigate the trade-off between nominal and robust performance, yielding state-of-the-art results on MNIST and CIFAR-10. Our code is available at: https://github.com/arobey1/advbench.

Accurate traffic state information plays a pivotal role in the Intelligent Transportation Systems (ITS), and it is an essential input to various smart mobility applications such as signal coordination and traffic flow prediction. The current practice to obtain the traffic state information is through specialized sensors such as loop detectors and speed cameras. In most metropolitan areas, traffic monitoring cameras have been installed to monitor the traffic conditions on arterial roads and expressways, and the collected videos or images are mainly used for visual inspection by traffic engineers. Unfortunately, the data collected from traffic monitoring cameras are affected by the 4L characteristics: Low frame rate, Low resolution, Lack of annotated data, and Located in complex road environments. Therefore, despite the great potentials of the traffic monitoring cameras, the 4L characteristics hinder them from providing useful traffic state information (e.g., speed, flow, density). This paper focuses on the traffic density estimation problem as it is widely applicable to various traffic surveillance systems. To the best of our knowledge, there is a lack of the holistic framework for addressing the 4L characteristics and extracting the traffic density information from traffic monitoring camera data. In view of this, this paper proposes a framework for estimating traffic density using uncalibrated traffic monitoring cameras with 4L characteristics. The proposed framework consists of two major components: camera calibration and vehicle detection. The camera calibration method estimates the actual length between pixels in the images and videos, and the vehicle counts are extracted from the deep-learning-based vehicle detection method. Combining the two components, high-granular traffic density can be estimated. To validate the proposed framework, two case studies were conducted in Hong Kong and Sacramento. The results show that the Mean Absolute Error (MAE) in camera calibration is less than 0.2 meters out of 6 meters, and the accuracy of vehicle detection under various conditions is approximately 90%. Overall, the MAE for the estimated density is 9.04 veh/km/lane in Hong Kong and 1.30 veh/km/lane in Sacramento. The research outcomes can be used to calibrate the speed-density fundamental diagrams, and the proposed framework can provide accurate and real-time traffic information without installing additional sensors.

Network load balancers are central components in data centers, that distributes workloads across multiple servers and thereby contribute to offering scalable services. However, when load balancers operate in dynamic environments with limited monitoring of application server loads, they rely on heuristic algorithms that require manual configurations for fairness and performance. To alleviate that, this paper proposes a distributed asynchronous reinforcement learning mechanism to - with no active load balancer state monitoring and limited network observations - improve the fairness of the workload distribution achieved by a load balancer. The performance of proposed mechanism is evaluated and compared with stateof-the-art load balancing algorithms in a simulator, under configurations with progressively increasing complexities. Preliminary results show promise in RLbased load balancing algorithms, and identify additional challenges and future research directions, including reward function design and model scalability.

This piece was the winner of the inaugural Gradient Prize. AI systems are compute-intensive: the AI lifecycle often requires long-running training jobs, hyperparameter searches, inference jobs, and other costly computations. They also require massive amounts of data that might be moved over the wire, and require specialized hardware to operate effectively, especially large-scale AI systems. All of these activities require electricity -- which has a carbon cost. There are also carbon emissions in ancillary needs like hardware and datacenter cooling [1]. Thus, AI systems have a massive carbon footprint[2].

Remitter, a fintech platform that improves bill payment rates by communicating to consumers digitally, has just finalised its acquisition of US-based Mercantile Adjustment Bureau, following a successful pre-IPO USD $12m cap raise led by Canaccord Genuity. The raise was cornerstoned by Allium Capital and Casey Capital. Remitter is a white-label communications platform, founded in Australia, which uses AI to optimise customer engagement and enhance the recovery of accounts receivables. Currently many organisations face challenges in collecting bill payments on time, with 46% of customers paying late according to Aite Group research. Remitter entered the US market in 2020, following two years of development including compliance across states and territories and collaboration with clients to ensure an optimal feature set. "Entering the US market involved ensuring the Remitter platform was compliant in all 52 states, each with its own different laws and regulations around payments.

Nominations are solicited for the 2022 ACM SIGAI Autonomous Agents Research Award. This award is made for excellence in research in the area of autonomous agents. It is intended to recognize researchers in autonomous agents whose current work is an important influence on the field. The award is an official ACM award, funded by an endowment created by ACM SIGAI from the proceeds of previous Autonomous Agents conferences. The recipient of the award will receive a monetary prize and a certificate, and will be invited to present a plenary talk at the AAMAS 2022 conference in Auckland, New Zealand.