To solve the conundrum of how to get timely medical care to people with memory loss or other impaired cognitive functioning, a new study suggests that artificial intelligence may be as accurate as clinicians in taking the first step: diagnosis. Findings from the study, which was conducted by researchers at Boston University School of Medicine, were published online Monday in the journal Nature Communications. "We're trying to leverage AI to create frameworks to mimic neurology experts," for dementia diagnosis, Vijaya B. Kolachalama, the study's principal investigator and assistant professor of medicine and computer science at Boston University, told UPI. He said his lab aims to use computer models to assist clinical practice. Kolachalama stressed that the aim of his team's work is to help reduce the workload of the busy neurology practice, not replace the expert clinician.

Recommender systems personalize content by recommending items to users. Item recommendation algorithms are evaluated by metrics that compare the positions of truly relevant items among the recommended items. To speed up the computation of metrics, recent work often uses sampled metrics where only a smaller set of random items and the relevant items are ranked. This paper investigates such sampled metrics in more detail and shows that they are inconsistent with their exact counterpart, in the sense that they do not persist relative statements, for example, recommender A is better than B, not even in expectation. Moreover, the smaller the sample size, the less difference there is between metrics, and for very small sample size, all metrics collapse to the AUC metric. We show that it is possible to improve the quality of the sampled metrics by applying a correction, obtained by minimizing different criteria. We conclude with an empirical evaluation of the naive sampled metrics and their corrected variants. To summarize, our work suggests that sampling should be avoided for metric calculation, however if an experimental study needs to sample, the proposed corrections can improve the quality of the estimate. Recommender systems are a key technology in online platforms for personalizing the selection of items that are shown to a user. Examples include recommending which products to buy, which videos to watch or which songs to play. Recommendations are typically user-dependent and often context-dependent. A key operation of recommender systems is to retrieve a ranked list of the best items for a user in a particular context.

Most real-world knowledge graphs (KG) are far from complete and comprehensive. This problem has motivated efforts in predicting the most plausible missing facts to complete a given KG, i.e., knowledge graph completion (KGC). However, existing KGC methods suffer from two main issues, 1) the false negative issue, i.e., the sampled negative training instances may include potential true facts; and 2) the data sparsity issue, i.e., true facts account for only a tiny part of all possible facts. To this end, we propose positive-unlabeled learning with adversarial data augmentation (PUDA) for KGC. In particular, PUDA tailors positive-unlabeled risk estimator for the KGC task to deal with the false negative issue. Furthermore, to address the data sparsity issue, PUDA achieves a data augmentation strategy by unifying adversarial training and positive-unlabeled learning under the positive-unlabeled minimax game. Extensive experimental results on real-world benchmark datasets demonstrate the effectiveness and compatibility of our proposed method.

Optimal transport (OT) is known to be sensitive against outliers because of its marginal constraints. Outlier robust OT variants have been proposed based on the definition that outliers are samples which are expensive to move. In this paper, we show that this definition is restricted by considering the case where outliers are closer to the target measure than clean samples. We show that outlier robust OT fully transports these outliers leading to poor performances in practice. To tackle these outliers, we propose to detect them by relying on a classifier trained with adversarial training to classify source and target samples. A sample is then considered as an outlier if the prediction from the classifier is different from its assigned label. To decrease the influence of these outliers in the transport problem, we propose to either remove them from the problem or to increase the cost of moving them by using the classifier prediction. We show that we successfully detect these outliers and that they do not influence the transport problem on several experiments such as gradient flows, generative models and label propagation.

We consider the problem of reaching a propositional goal condition in fully-observable nondeterministic (FOND) planning under a general class of fairness assumptions that are given explicitly. The fairness assumptions are of the form A/B and say that state trajectories that contain infinite occurrences of an action a from A in a state s and finite occurrence of actions from B, must also contain infinite occurrences of action a in s followed by each one of its possible outcomes. The infinite trajectories that violate this condition are deemed as unfair, and the solutions are policies for which all the fair trajectories reach a goal state. We show that strong and strong-cyclic FOND planning, as well as QNP planning, a planning model introduced recently for generalized planning, are all special cases of FOND planning with fairness assumptions of this form which can also be combined. FOND+ planning, as this form of planning is called, combines the syntax of FOND planning with some of the versatility of LTL for expressing fairness constraints. A sound and complete FOND+ planner is implemented by reducing FOND+ planning to answer set programs, and its performance is evaluated in comparison with FOND and QNP planners, and LTL synthesis tools. Two other FOND+ planners are introduced as well which are more scalable but are not complete.  

Shap explanations are a popular feature-attribution mechanism for explainable AI. They use game-theoretic notions to measure the influence of individual features on the prediction of a machine learning model. Despite a lot of recent interest from both academia and industry, it is not known whether Shap explanations of common machine learning models can be computed efficiently. In this paper, we establish the complexity of computing the Shap explanation in three important settings. First, we consider fully-factorized data distributions, and show that the complexity of computing the Shap explanation is the same as the complexity of computing the expected value of the model. This fully-factorized setting is often used to simplify the Shap computation, yet our results show that the computation can be intractable for commonly used models such as logistic regression. Going beyond fully-factorized distributions, we show that computing Shap explanations is already intractable for a very simple setting: computing Shap explanations of trivial classifiers over naive Bayes distributions. Finally, we show that even computing Shap over the empirical distribution is #P-hard.

Last year, researchers from Curtin University tracked the fall of a meteorite in Western Australia, using advanced technologies.

Researchers have warned against possible harms from AI that processes and generates text.Credit: Getty An international team of around 1,000 largely academic volunteers has tried to break big tech's stranglehold on natural-language processing and reduce its harms. Trained with US$7-million-worth of publicly funded computing time, the BLOOM language model will rival in scale those made by firms Google and OpenAI, but will be open-source. BLOOM will also be the first model of its scale to be multilingual. The collaboration, called BigScience, launched an early version of the model on 17 June, and hopes that it will ultimately help to reduce harmful outputs of artificial intelligence (AI) language systems. Models that recognize and generate language are increasingly used by big tech firms in applications from chat bots to translators, and can sound so eerily human that a Google engineer this month claimed that the firm's AI model was sentient (Google strongly denies that the AI possesses sentience).

Bilevel optimization have gained growing interests, with numerous applications found in meta learning, minimax games, reinforcement learning, and nested composition optimization. This paper studies the problem of distributed bilevel optimization over a network where agents can only communicate with neighbors, including examples from multi-task, multi-agent learning and federated learning. In this paper, we propose a gossip-based distributed bilevel learning algorithm that allows networked agents to solve both the inner and outer optimization problems in a single timescale and share information via network propagation. We show that our algorithm enjoys the $\mathcal{O}(\frac{1}{K \epsilon^2})$ per-agent sample complexity for general nonconvex bilevel optimization and $\mathcal{O}(\frac{1}{K \epsilon})$ for strongly convex objective, achieving a speedup that scales linearly with the network size. The sample complexities are optimal in both $\epsilon$ and $K$. We test our algorithm on the examples of hyperparameter tuning and decentralized reinforcement learning. Simulated experiments confirmed that our algorithm achieves the state-of-the-art training efficiency and test accuracy.

It is common in computer vision to be confronted with domain shift: images which have the same class but different acquisition conditions. In domain adaptation (DA), one wants to classify unlabeled target images using source labeled images. Unfortunately, deep neural networks trained on a source training set perform poorly on target images which do not belong to the training domain. One strategy to improve these performances is to align the source and target image distributions in an embedded space using optimal transport (OT). However OT can cause negative transfer, i.e. aligning samples with different labels, which leads to overfitting especially in the presence of label shift between domains. In this work, we mitigate negative alignment by explaining it as a noisy label assignment to target images. We then mitigate its effect by appropriate regularization. We propose to couple the MixUp regularization \citep{zhang2018mixup} with a loss that is robust to noisy labels in order to improve domain adaptation performance. We show in an extensive ablation study that a combination of the two techniques is critical to achieve improved performance. Finally, we evaluate our method, called \textsc{mixunbot}, on several benchmarks and real-world DA problems.