The application of machine learning to support the processing of large datasets holds promise in many industries, including financial services. However, practical issues for the full adoption of machine learning remain with the focus being on understanding and being able to explain the decisions and predictions made by complex models. In this paper, we explore explainability methods in the domain of real-time fraud detection by investigating the selection of appropriate background datasets and runtime trade-offs on both supervised and unsupervised models.

Machine unlearning has great significance in guaranteeing model security and protecting user privacy. Additionally, many legal provisions clearly stipulate that users have the right to demand model providers to delete their own data from training set, that is, the right to be forgotten. The naive way of unlearning data is to retrain the model without it from scratch, which becomes extremely time and resource consuming at the modern scale of deep neural networks. Other unlearning approaches by refactoring model or training data struggle to gain a balance between overhead and model usability. In this paper, we propose an approach, dubbed as DeepObliviate, to implement machine unlearning efficiently, without modifying the normal training mode. Our approach improves the original training process by storing intermediate models on the hard disk. Given a data point to unlearn, we first quantify its temporal residual memory left in stored models. The influenced models will be retrained and we decide when to terminate the retraining based on the trend of residual memory on-the-fly. Last, we stitch an unlearned model by combining the retrained models and uninfluenced models. We extensively evaluate our approach on five datasets and deep learning models. Compared to the method of retraining from scratch, our approach can achieve 99.0%, 95.0%, 91.9%, 96.7%, 74.1% accuracy rates and 66.7$\times$, 75.0$\times$, 33.3$\times$, 29.4$\times$, 13.7$\times$ speedups on the MNIST, SVHN, CIFAR-10, Purchase, and ImageNet datasets, respectively. Compared to the state-of-the-art unlearning approach, we improve 5.8% accuracy, 32.5$\times$ prediction speedup, and reach a comparable retrain speedup under identical settings on average on these datasets. Additionally, DeepObliviate can also pass the backdoor-based unlearning verification.

Questions of fairness, robustness, and transparency are paramount to address before deploying NLP systems. Central to these concerns is the question of reliability: Can NLP systems reliably treat different demographics fairly and function correctly in diverse and noisy environments? To address this, we argue for the need for reliability testing and contextualize it among existing work on improving accountability. We show how adversarial attacks can be reframed for this goal, via a framework for developing reliability tests. We argue that Figure 1: How DOCTOR can integrate with existing reliability testing -- with an emphasis on interdisciplinary system development workflows. Test (left) and system collaboration -- will enable rigorous development (right) take place in parallel, separate and targeted testing, and aid in the enactment teams. Reliability tests can thus be constructed independent and enforcement of industry standards. of the system development team, either by an internal "red team" or by independent auditors.

Advances in AI are making their way across all products and services we interact with. Our cars are outfitted with tools that trigger automatic breaking, platforms such as Netflix proactively suggest recommendations for viewing, Alexa and Google can predict our search needs, and Spotify can recommend songs and curate listening lists much better than you or I can. Although the advantages of AI in our daily lives are undeniable, people are concerned about its dangers. Inadequate physical security, economic losses, and ethical issues are just a few examples of the damage AI could cause. In response to AI dangers, the European Union is working on a legal framework to regulate artificial intelligence.

UNESCO 2019 publication I d Blush if I Could revealed how much gender bias and stereotypes were engineered into Artificial Intelligence-powered voice-assistant applications. Beyond highlighting the overall gender imbalance of teams creating these new tools, it also showed evidence of alarming gender gaps in technology industries, even in countries that are close to achieving gender equality. In this Q&A, Mark West, Project Officer at UNESCO and lead-author of the publication, shares insights on where we stand regarding gender prejudice in AI since the report came out. Two years since the publication came out, where do we stand on the fight against gender prejudice in AI assistants? On the positive side, awareness is much higher than it was when we researched and wrote our report.

Japan's parliament on Wednesday enacted a set of laws to establish a new government agency in September as the country aims to speed up digitalization. Prime Minister Yoshihide Suga's government hopes to accelerate digitalization in the central and local governments to improve the quality of services after the novel coronavirus pandemic exposed challenges caused by a delay in the initiatives. Under the digitalization legislation that was enacted in the Upper House plenary session, Japan will revamp computer systems for central and local governments and introduce common nationwide rules to protect personal information. It will also be tasked with taking privacy protection measures as the legislation will boost the exchange of personal information, in a move that could cause data leakage and other risks. The Suga government has, since its launch last September, placed a high priority on digital reforms, as Japan has long been struggling to promote administrative reforms by utilizing information technology, despite having aimed at improvements since around 2000.

The relevance of the Key Information Extraction (KIE) task is increasingly important in natural language processing problems. But there are still only a few well-defined problems that serve as benchmarks for solutions in this area. To bridge this gap, we introduce two new datasets (Kleister NDA and Kleister Charity). They involve a mix of scanned and born-digital long formal English-language documents. In these datasets, an NLP system is expected to find or infer various types of entities by employing both textual and structural layout features. The Kleister Charity dataset consists of 2,788 annual financial reports of charity organizations, with 61,643 unique pages and 21,612 entities to extract. The Kleister NDA dataset has 540 Non-disclosure Agreements, with 3,229 unique pages and 2,160 entities to extract. We provide several state-of-the-art baseline systems from the KIE domain (Flair, BERT, RoBERTa, LayoutLM, LAMBERT), which show that our datasets pose a strong challenge to existing models. The best model achieved an 81.77% and an 83.57% F1-score on respectively the Kleister NDA and the Kleister Charity datasets. We share the datasets to encourage progress on more in-depth and complex information extraction tasks.

Recent advances in artificial intelligence (AI) have achieved human-scale speed and accuracy for classification tasks. In turn, these capabilities have made AI a viable replacement for many human activities that at their core involve classification, such as basic mechanical and analytical tasks in low-level service jobs. Current systems do not need to be conscious to recognize patterns and classify them. However, for AI to progress to more complicated tasks requiring intuition and empathy, it must develop capabilities such as metathinking, creativity, and empathy akin to human self-awareness or consciousness. We contend that such a paradigm shift is possible only through a fundamental shift in the state of artificial intelligence toward consciousness, a shift similar to what took place for humans through the process of natural selection and evolution. As such, this paper aims to theoretically explore the requirements for the emergence of consciousness in AI. It also provides a principled understanding of how conscious AI can be detected and how it might be manifested in contrast to the dominant paradigm that seeks to ultimately create machines that are linguistically indistinguishable from humans.

Generating accurate terminology is a crucial component for the practicality and reliability of neural machine translation (NMT) systems. To address this, lexically constrained NMT explores various methods to ensure pre-specified words and phrases to appear in the translations. In many cases, however, those methods are evaluated on general domain corpora, where the terms are mostly uni- and bi-grams (>98%). In this paper, we instead tackle a more challenging setup consisting of domain-specific corpora with much longer n-gram and highly specialized terms. To encourage span-level representations in generation, we additionally impose a source-sentence conditioned masked span prediction loss in the decoder and observe improvements on both terminology translation as well as BLEU scores. Experimental results on three domain-specific corpora in two language pairs demonstrate that the proposed training scheme can improve the performance of existing lexically constrained methods that can operate both with or without a term dictionary at test time.

Defeasible reasoning is the mode of reasoning where conclusions can be overturned by taking into account new evidence. A commonly used method in philosophy and AI literature is to handcraft argumentation supporting inference graphs. While humans find inference graphs very useful for reasoning, constructing them at scale is difficult. In this paper, we automatically generate such inference graphs through transfer learning from another NLP task that shares the kind of reasoning that inference graphs support. Through automated metrics and human evaluation, we find that our method generates meaningful graphs for the defeasible inference task. Human accuracy on this task improves by 20% by consulting the generated graphs. Our findings open up exciting new research avenues for cases where machine reasoning can help human reasoning. (A dataset of 230,000 influence graphs for each defeasible query is located at: https://tinyurl.com/defeasiblegraphs.)