While neural networks are good at learning unspecified functions from training samples, they cannot be directly implemented in hardware and are often not interpretable or formally verifiable. On the other hand, logic circuits are implementable, verifiable, and interpretable but are not able to learn from training data in a generalizable way. We propose a novel logic learning pipeline that combines the advantages of neural networks and logic circuits. Our pipeline first trains a neural network on a classification task, and then translates this, first to random forests or look-up tables, and then to AND-Inverter logic. We show that our pipeline maintains greater accuracy than naive translations to logic, and minimizes the logic such that it is more interpretable and has decreased hardware cost. We show the utility of our pipeline on a network that is trained on biomedical data from patients presenting with gastrointestinal bleeding with the prediction task of determining if patients need immediate hospital-based intervention. This approach could be applied to patient care to provide risk stratification and guide clinical decision-making.

Emulators speed up simulations, such as this NASA aerosol model that shows soot from fires in Australia. Modeling immensely complex natural phenomena such as how subatomic particles interact or how atmospheric haze affects climate can take many hours on even the fastest supercomputers. Emulators, algorithms that quickly approximate these detailed simulations, offer a shortcut. Now, work posted online shows how artificial intelligence (AI) can easily produce accurate emulators that can accelerate simulations across all of science by billions of times. "This is a big deal," says Donald Lucas, who runs climate simulations at Lawrence Livermore National Laboratory and was not involved in the work.

One of the best performers on the ASX on Monday was the BrainChip Holdings Ltd (ASX: BRN) share price. The artificial intelligence company's shares rocketed 25% higher to 6.9 cents at one stage before closing the day 14.5% higher. Investors were buying the company's shares after it announced the receipt of an EAR99 classification for its Akida Neuromorphic System-on-Chip (NSoC), Akida Software Development Environment (ADE), and related technologies from the U.S. Government. The Export Administration Regulations (EAR) classification of EAR99, which BrainChip has now formally received, removes the barriers for exporting Akida to non-U.S. The EAR99 designation means the company does not require a pre-approval, or a license from the U.S. Department of Commerce, before delivering its solutions globally as part of sales and market expansion activities.

The current understanding of deep neural networks can only partially explain how input structure, network parameters and optimization algorithms jointly contribute to achieve the strong generalization power that is typically observed in many real-world applications. In order to improve the comprehension and interpretability of deep neural networks, we here introduce a novel theoretical framework based on the compositional structure of piecewise linear activation functions. By defining a direct acyclic graph representing the composition of activation patterns through the network layers, it is possible to characterize the instances of the input data with respect to both the predicted label and the specific (linear) transformation used to perform predictions. Preliminary tests on the MNIST dataset show that our method can group input instances with regard to their similarity in the internal representation of the neural network, providing an intuitive measure of input complexity.

We develop a chatbot using Deep Bidirectional Transformer models (BERT) to handle client questions in financial investment customer service. The bot can recognize 381 intents, and decides when to say "I don't know" and escalates irrelevant/uncertain questions to human operators. Our main novel contribution is the discussion about uncertainty measure for BERT, where three different approaches are systematically compared on real problems. We investigated two uncertainty metrics, information entropy and variance of dropout sampling in BERT, followed by mixed-integer programming to optimize decision thresholds. Another novel contribution is the usage of BERT as a language model in automatic spelling correction. Inputs with accidental spelling errors can significantly decrease intent classification performance. The proposed approach combines probabilities from masked language model and word edit distances to find the best corrections for misspelled words. The chatbot and the entire conversational AI system are developed using open-source tools, and deployed within our company's intranet. The proposed approach can be useful for industries seeking similar in-house solutions in their specific business domains. We share all our code and a sample chatbot built on a public dataset on Github.

Most neural networks utilize the same amount of compute for every example independent of the inherent complexity of the input. Further, methods that adapt the amount of computation to the example focus on finding a fixed inference-time computational graph per example, ignoring any external computational budgets or varying inference time limitations. In this work, we utilize conditional computation to make neural sequence models (Transformer) more efficient and computation-aware during inference. We first modify the Transformer architecture, making each set of operations conditionally executable depending on the output of a learned control network. We then train this model in a multi-task setting, where each task corresponds to a particular computation budget. This allows us to train a single model that can be controlled to operate on different points of the computation-quality trade-off curve, depending on the available computation budget at inference time. We evaluate our approach on two tasks: (i) WMT English-French Translation and (ii) Unsupervised representation learning (BERT). Our experiments demonstrate that the proposed Conditional Computation Transformer (CCT) is competitive with vanilla Transformers when allowed to utilize its full computational budget, while improving significantly over computationally equivalent baselines when operating on smaller computational budgets.

In this work, we take a closer look at the evaluation of two families of methods for enriching information from knowledge graphs: Link Prediction and Entity Alignment. In the current experimental setting, multiple different scores are employed to assess different aspects of model performance. We analyze the informative value of these evaluation measures and identify several shortcomings. In particular, we demonstrate that all existing scores can hardly be used to compare results across different datasets. Moreover, this problem may also arise when comparing different train/test splits for the same dataset. We show that this leads to various problems in the interpretation of results, which may support misleading conclusions. Therefore, we propose a different evaluation and demonstrate empirically how this helps for fair, comparable and interpretable assessment of model performance.

We study an online learning problem subject to the constraint of individual fairness, which requires that similar individuals are treated similarly. Unlike prior work on individual fairness, we do not assume the similarity measure among individuals is known, nor do we assume that such measure takes a certain parametric form. Instead, we leverage the existence of an auditor who detects fairness violations without enunciating the quantitative measure. In each round, the auditor examines the learner's decisions and attempts to identify a pair of individuals that are treated unfairly by the learner. We provide a general reduction framework that reduces online classification in our model to standard online classification, which allows us to leverage existing online learning algorithms to achieve sub-linear regret and number of fairness violations. Surprisingly, in the stochastic setting where the data are drawn independently from a distribution, we are also able to establish PAC-style fairness and accuracy generalization guarantees (Yona and Rothblum [2018]), despite only having access to a very restricted form of fairness feedback. Our fairness generalization bound qualitatively matches the uniform convergence bound of Yona and Rothblum [2018], while also providing a meaningful accuracy generalization guarantee. Our results resolve an open question by Gillen et al. [2018] by showing that online learning under an unknown individual fairness constraint is possible even without assuming a strong parametric form of the underlying similarity measure.

Artificial Intelligence (AI) systems that companies claim can "read" facial expressions is based on outdated science and risks being unreliable and discriminatory, one of the world's leading experts on the psychology of emotion has warned. Lisa Feldman Barrett, professor of psychology at Northeastern University, said that such technologies appear to disregard a growing body of evidence undermining the notion that the basic facial expressions are universal across cultures. As a result, such technologies – some of which are already being deployed in real-world settings – run the risk of being unreliable or discriminatory, she said. "I don't know how companies can continue to justify what they're doing when it's really clear what the evidence is," she said. "There are some companies that just continue to claim things that can't possibly be true."

China's Communist Government has extracted over 6 billion peoples biometrics, including facial, voice and personal health data to empower their Quantum Artificial Intelligence program meant for military purposes. This includes almost every American, Canadian, and European persons living today, every person in China, and Less so from groups in Africa, the Middle East, and South America. I initially made the finding public by publishing the discovery in the book AI, Trump, China and the Weaponization of Robotics without providing company names. Later, I included the findings with company names in the updated book Artificial Intelligence Dangers to Humanity. More than 1,000 AI, Robotics and Bio-Metric companies were researched to obtain the results of over 6 billion human beings who have had their bio-metrics stolen or transferred to China.