The learner provides an unlabeled example to the oracle, and the oracle responds with the label.

Paper proposes a new method for time-sensitive recommendations based on user activities. It is different from existing methods because it address the problem of time-varying user preferences. The paper also addresses prediction of the next returning time of a user. Learning is performed using an efficient optimization algorithm proposed by authors. Experimental section shows results on one synthetic data set in order to show that learning is efficient on large scale data and on two real data sets of modest size.

Markov jump processes are continuous-time stochastic processes which describe dynamical systems evolving in discrete state spaces. These processes find wide application in the natural sciences and machine learning, but their inference is known to be far from trivial. In this work we introduce a methodology for zeroshot inference of Markov jump processes (MJPs), on bounded state spaces, from noisy and sparse observations, which consists of two components. First, a broad probability distribution over families of MJPs, as well as over possible observation times and noise mechanisms, with which we simulate a synthetic dataset of hidden MJPs and their noisy observation process. Second, a neural network model that processes subsets of the simulated observations, and that is trained to output the initial condition and rate matrix of the target MJP in a supervised way. We empirically demonstrate that one and the same (pretrained) model can infer, in a zero-shot fashion, hidden MJPs evolving in state spaces of different dimensionalities. Specifically, we infer MJPs which describe (i) discrete flashing ratchet systems, which are a type of Brownian motors, and the conformational dynamics in (ii) molecular simulations, (iii) experimental ion channel data and (iv) simple protein folding models. What is more, we show that our model performs on par with state-of-the-art models which are finetuned to the target datasets. Our pretrained model is available online.

The learner provides an unlabeled example to the oracle, and the oracle responds with the label.

Jupyter Notebook is an interactive development environment commonly used for rapid experimentation of machine learning (ML) solutions. Describing the ML activities performed along code cells improves the readability and understanding of Notebooks. Manual annotation of code cells is time-consuming and error-prone. Therefore, tools have been developed that classify the cells of a notebook concerning the ML activity performed in them. However, the current tools are not flexible, as they work based on look-up tables that have been created, which map function calls of commonly used ML libraries to ML activities. These tables must be manually adjusted to account for new or changed libraries. This paper presents a more flexible approach to cell classification based on a hybrid classification approach that combines a rule-based and a decision tree classifier. We discuss the design rationales and describe the developed classifiers in detail. We implemented the new flexible cell classification approach in a tool called JupyLabel. Its evaluation and the obtained metric scores regarding precision, recall, and F1-score are discussed. Additionally, we compared JupyLabel with HeaderGen, an existing cell classification tool. We were able to show that the presented flexible cell classification approach outperforms this tool significantly.

Neural 'dense' retrieval models are state of the art for many datasets, however these models often exhibit limited domain transfer ability. Existing approaches to adaptation are unwieldy, such as requiring explicit supervision, complex model architectures, or massive external models. We present $\texttt{ABEL}$, a simple but effective unsupervised method to enhance passage retrieval in zero-shot settings. Our technique follows a straightforward loop: a dense retriever learns from supervision signals provided by a reranker, and subsequently, the reranker is updated based on feedback from the improved retriever. By iterating this loop, the two components mutually enhance one another's performance. Experimental results demonstrate that our unsupervised $\texttt{ABEL}$ model outperforms both leading supervised and unsupervised retrievers on the BEIR benchmark. Meanwhile, it exhibits strong adaptation abilities to tasks and domains that were unseen during training. By either fine-tuning $\texttt{ABEL}$ on labelled data or integrating it with existing supervised dense retrievers, we achieve state-of-the-art results.\footnote{Source code is available at \url{https://github.com/Fantabulous-J/BootSwitch}.}

While recent works have been considerably improving the quality of the natural language explanations (NLEs) generated by a model to justify its predictions, there is very limited research in detecting and alleviating inconsistencies among generated NLEs. In this work, we leverage external knowledge bases to significantly improve on an existing adversarial attack for detecting inconsistent NLEs. We apply our attack to high-performing NLE models and show that models with higher NLE quality do not necessarily generate fewer inconsistencies. Moreover, we propose an off-the-shelf mitigation method to alleviate inconsistencies by grounding the model into external background knowledge. Our method decreases the inconsistencies of previous high-performing NLE models as detected by our attack.

'Gutfeld!' panelists weigh in on a new survey claiming almost half of recent college graduates aren't'emotionally' prepared for a 9-5 job. Student senator Mike Abel at the University of Houston successfully challenged the Student Government Association's (SGA) "Respect for Pronouns" bill that would have mandated the use of others members' preferred pronouns. The bill also calls for "Name tags containing the proper pronouns will be given to every member of the Student Government Association" and it "strongly recommend[ed]" to list pronouns during Zoom meetings. The SGA's Supreme Court is poised to rule in favor of Abel, determining that such legislation constituted a violation of the students' first amendment rights, Campus Reform reported Friday. Specifically, the SGA Supreme Court is said to have found that the last sentence of the legislation, which would have compelled speech from the organization's members, was a violation.

AI is helping to diagnose tuberculosis and cancer more accurately. Reports suggest that there was a 33 per cent increase in the notification rate, and the number of drop-outs of presumptive cases fell from 72 per cent to 53 per cent. The applications and use-cases are innumerable, and increasingly, the Indian health care scene is warming to AI-based diagnoses and interventions. Nilesh Shah, president and chief of science and innovation at Metropolis Healthcare, says that several common tests such as complete blood count and even tests for autoimmune disorders can be done quickly and without errors using AI-enabled applications. "The share of AI in diagnosis is only going to go up in future, and this will reduce the burden on doctors, who can devote their time to complex cases. Also, there is a huge cost benefit," says Shah.

In short, the Internet of Things, big data and Artificial Intelligence is being used to interpret the language of bees to gain a true understanding of biodiversity and environmental health. "We're starting to understand the characteristics of communication in the beehive," says John Abel, vice president of cloud and technology at Oracle for the UK, Ireland and Israel. "Already, we understand certain actions of the bee. For example, flying in a figure 8 is not random. Certain tones bees use will indicate food or water. The way the bee shudders and rotates within the figure 8 will indicate to the rest of the colony what it found and where it is. If the heat or sound in the hive changes, it can mean the hive is preparing to swarm. "If the queen bee is too large to fly – because when it is in the hive, its job is to create the future bees of the hive – the queen's workers have to prepare the queen for flying, and that takes 20 odd days.