This thesis presents new methods for unsupervised learning of distributed representations of words and entities from text and knowledge bases. The first algorithm presented in the thesis is a multi-view algorithm for learning representations of words called Multiview Latent Semantic Analysis (MVLSA). By incorporating up to 46 different types of co-occurrence statistics for the same vocabulary of english words, I show that MVLSA outperforms other state-of-the-art word embedding models. Next, I focus on learning entity representations for search and recommendation and present the second method of this thesis, Neural Variational Set Expansion (NVSE). NVSE is also an unsupervised learning method, but it is based on the Variational Autoencoder framework. Evaluations with human annotators show that NVSE can facilitate better search and recommendation of information gathered from noisy, automatic annotation of unstructured natural language corpora. Finally, I move from unstructured data and focus on structured knowledge graphs. I present novel approaches for learning embeddings of vertices and edges in a knowledge graph that obey logical constraints.

Understanding temporal and causal relations between events is a fundamental natural language understanding task. Because a cause must be before its effect in time, temporal and causal relations are closely related and one relation even dictates the other one in many cases. However, limited attention has been paid to studying these two relations jointly. This paper presents a joint inference framework for them using constrained conditional models (CCMs). Specifically, we formulate the joint problem as an integer linear programming (ILP) problem, enforcing constraints inherently in the nature of time and causality. We show that the joint inference framework results in statistically significant improvement in the extraction of both temporal and causal relations from text.

Obtaining training data for Question Answering (QA) is time-consuming and resource-intensive, and existing QA datasets are only available for limited domains and languages. In this work, we explore to what extent high quality training data is actually required for Extractive QA, and investigate the possibility of unsupervised Extractive QA. We approach this problem by first learning to generate context, question and answer triples in an unsupervised manner, which we then use to synthesize Extractive QA training data automatically. To generate such triples, we first sample random context paragraphs from a large corpus of documents and then random noun phrases or named entity mentions from these paragraphs as answers. Next we convert answers in context to "fill-in-the-blank" cloze questions and finally translate them into natural questions. We propose and compare various unsupervised ways to perform cloze-to-natural question translation, including training an unsupervised NMT model using non-aligned corpora of natural questions and cloze questions as well as a rule-based approach. We find that modern QA models can learn to answer human questions surprisingly well using only synthetic training data. We demonstrate that, without using the SQuAD training data at all, our approach achieves 56.4 F1 on SQuAD v1 (64.5 F1 when the answer is a Named entity mention), outperforming early supervised models.

Rafael, an Israeli defence firm, on Monday announced it had successfully demonstrated a new "automatic target recognition" capability that relies on artificial intelligence and machine learning for the newest variant of its SPICE family of guided air-to-ground bombs. The Indian Air Force is believed to have used SPICE bombs in its attack on a Jaish-e-Mohammed camp in Balakot in February. SPICE is an acronym for Smart, Precise, Impact and Cost-Effective. The SPICE munitions come in three variants: SPICE-2000, -1000 and -250, with the number denoting the weapon's weight class in pounds. The SPICE-250, which weighs around 113kg and the newest variant of the SPICE family, was the version tested with artificial intelligence technology by Rafael.

The American Medical Informatics Association wants the Food and Drug Administration to improve its conceptual approach to regulating medical devices that leverage self-updating artificial intelligence algorithms. The FDA sees tremendous potential in healthcare for AI algorithms that continually evolve--called "adaptive" or "continuously learning" algorithms--that don't need manual modification to incorporate learning or updates. While AMIA supports an FDA discussion paper on the topic released in early April, the group is calling on the agency to make further refinements to the Proposed Regulatory Framework for Modifications to Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD). "Properly regulating AI and machine learning-based SaMD will require ongoing dialogue between FDA and stakeholders," said AMIA President and CEO Douglas Fridsma, MD, in a written statement. "This draft framework is only the beginning of a vital conversation to improve both patient safety and innovation. We certainly look forward to continuing it."

Perth City council is pressing ahead with a trial of facial recognition technology to be installed in cameras across East Perth, despite concerns from privacy experts and local residents. The network of 30 cameras is set to go live within weeks, amid complaints there has been no proper local consultation since the plans were revealed last year. The cameras are equipped with software that uses deep-learning AI to recognise faces, count passing people and vehicles and track movement. The council said the system was going through final tests, and was expected to be activated within a few weeks. The secretary of the East Perth Community Safety Group, Lyn Schwan, said she was not aware of any community consultation about the trial.

The artificial intelligence (AI) landscape has evolved significantly from 1950 when Alan Turing first posed the question of whether machines can think. Today, AI is transforming societies and economies. It promises to generate productivity gains, improve wellbeing and help address global challenges, such as climate change, resource scarcity and health crises. Yet, as AI applications are adopted around the world, their use can raises questions and challenges related to human values, fairness, human determination, privacy, safety and accountability, among others. This report helps build a shared understanding of AI in the present and near-term by mapping the AI technical, economic, use case and policy landscape and identifying major public policy considerations.

Just because we can use it, should we? That's the question more and more people are asking about face recognition technology, software that's already in our phones and our social media feeds and many security systems. San Francisco leaders have voted to ban the police from using it, and even some in the tech industry say there should be limits. BRAD SMITH: It's the kind of technology that can do a lot of good for a lot of people, but it can be misused. It can be used in ways that lead to discrimination and bias.

The news that the Home Office is sorting applications for visas with secret algorithms applied to online applications is a reminder of one of Theresa May's more toxic and long-lasting legacies: her immigration policies as home secretary. Yet even if the government's aims in immigration policy were fair and balanced, there would still be serious issues of principle involved in digitising the process. Handing over life-changing decisions to machine-learning algorithms is always risky. Small biases in the data become large biases in the outcome, but these are difficult to challenge because the use of software shrouds them in clouds of obfuscation and supposed objectivity, especially when its workings are described as "artificial intelligence". This is not to say they are always harmful, or never any use: with careful training and well-understood, clearly defined problems, and when they are operating on good data, software systems can perform much better than humans ever could.

We welcome written submissions from individuals and organisations who are developing policy, systems or safeguards on the use of AI as we gather evidence for this review. Please contact the Committee at public@public-standards.gov.uk to express your interest. Please see the review's Terms of Reference here. Read the Committee's blog post'why CSPL are reviewing artificial intelligence in the public sector' here.