The COVID-19 pandemic has demonstrated the increasing need of policymakers for timely estimates of macroeconomic variables. A prior UNCTAD research paper examined the suitability of long short-term memory artificial neural networks (LSTM) for performing economic nowcasting of this nature. Here, the LSTM's performance during the COVID-19 pandemic is compared and contrasted with that of the dynamic factor model (DFM), a commonly used methodology in the field. Three separate variables, global merchandise export values and volumes and global services exports, were nowcast with actual data vintages and performance evaluated for the second, third, and fourth quarters of 2020 and the first and second quarters of 2021. In terms of both mean absolute error and root mean square error, the LSTM obtained better performance in two-thirds of variable/quarter combinations, as well as displayed more gradual forecast evolutions with more consistent narratives and smaller revisions. Additionally, a methodology to introduce interpretability to LSTMs is introduced and made available in the accompanying nowcast_lstm Python library, which is now also available in R, MATLAB, and Julia.

In an out-of-distribution (OOD) detection problem, samples of known classes(also called in-distribution classes) are used to train a special classifier. In testing, the classifier can (1) classify the test samples of known classes to their respective classes and also (2) detect samples that do not belong to any of the known classes (i.e., they belong to some unknown or OOD classes). This paper studies the problem of zero-shot out-of-distribution(OOD) detection, which still performs the same two tasks in testing but has no training except using the given known class names. This paper proposes a novel yet simple method (called ZOC) to solve the problem. ZOC builds on top of the recent advances in zero-shot classification through multi-modal representation learning. It first extends the pre-trained language-vision model CLIP by training a text-based image description generator on top of CLIP. In testing, it uses the extended model to generate candidate unknown class names for each test sample and computes a confidence score based on both the known class names and candidate unknown class names for zero-shot OOD detection. Experimental results on 5 benchmark datasets for OOD detection demonstrate that ZOC outperforms the baselines by a large margin.

Governments and humanitarian groups can use machine learning algorithms and mobile phone data to get aid to those who need it most during a humanitarian crisis, we found in new research. The simple idea behind this approach, as we explained in the journal Nature on March 16, 2022, is that wealthy people use phones differently from poor people. Their phone calls and text messages follow different patterns, and they use different data plans, for example. Machine learning algorithms--which are fancy tools for pattern recognition--can be trained to recognize those differences and infer whether a given mobile subscriber is wealthy or poor. As the COVID-19 pandemic spread in early 2020, our research team helped Togo's Ministry of Digital Economy and GiveDirectly, a nonprofit that sends cash to people living in poverty, turn this insight into a new type of aid program. First, we collected recent, reliable and representative data.

March 17, 2022--Moshe Shenfeld, a computer science Ph.D. candidate at Hebrew University of Jerusalem (HU)'s Rachel and Selim Benin School of Engineering and Computer Science, was selected as an Apple Scholar in AI/Machine Learning for 2022. Shenfeld is one of only 15 awardees worldwide, the other Israeli recipient is from Tel Aviv University. The Ph.D. fellowship in Machine Learning and AI was created by Apple "to celebrate the contributions of students pursuing cutting-edge fundamental and applied machine learning research worldwide." Currently, Shenfeld is researching privacy-preserving machine learning under the supervision of HU Professor Katrina Ligett. His Ph.D. focuses on differential privacy and its relation to adaptive data analysis and machine learning.

The malaria parasite, which is transmitted by several Anopheles mosquito species, requires more time to reach its human-transmissible stage than the average lifespan of mosquito vectors. Monitoring the species-specific age structure of mosquito populations is critical to evaluating the impact of vector control interventions on malaria risk. We present a rapid, cost-effective surveillance method based on deep learning of mid-infrared spectra of mosquito cuticle that simultaneously identifies the species and age class of three main malaria vectors in natural populations. Using spectra from over 40, 000 ecologically and genetically diverse An. gambiae, An. arabiensis, and An. coluzzii females, we develop a deep transfer learning model that learns and predicts the age of new wild populations in Tanzania and Burkina Faso with minimal sampling effort. Additionally, the model is able to detect the impact of simulated control interventions on mosquito populations, measured as a shift in their age structures. In the future, we anticipate our method can be applied to other arthropod vector-borne diseases. Knowing the age of malaria-transmitting mosquitoes is important to understand transmission risk as only old mosquitoes can transmit the disease. Here, the authors develop a method based on mid-infrared spectra of mosquito cuticle that can rapidly identify the species and age class of main malaria vectors.

Much has been said about the potential of artificial intelligence (AI) to transform how we live, work, and interact with each other. But we must also draw attention to a less discussed, but equally important, question -- do we have the skills required to develop AI inclusively and use it responsibly? AI adoption is accelerating, and the overall market is expected to be worth $190 billion by 2025. By 2030, AI technology will add $15.7 trillion to global gross domestic product (GDP). AI is everywhere -- whether we're aware of it or not.

Speaking behind a podium bearing the Ukrainian state emblem, President Volodymyr Zelenskiy, in his now signature green attire, calls on his soldiers to lay down their weapons and return to their families. The one-minute clip is a deepfake, the term for a sophisticated hoax that uses artificial intelligence to create a phoney image, most commonly fake videos of people. A deepfake of Ukrainian President Volodymyr Zelensky calling on his soldiers to lay down their weapons was reportedly uploaded to a hacked Ukrainian news website today, per @Shayan86 pic.twitter.com/tXLrYECGY4 What unfolded next was the latest episode in the infowar that has accompanied the Russia-Ukraine conflict, a war being waged across social media platforms, via satellite images of battlefields and on hackers' keyboards. Zelenskiy posted a bona fide response on his Instagram account on Wednesday dismissing the "childish provocation" and telling Russian troops to return home.

Recently, prefix-tuning has gained increasing attention as a parameter-efficient finetuning method for large-scale pretrained language models. The method keeps the pretrained models fixed and only updates the prefix token parameters for each downstream task. Despite being lightweight and modular, prefix-tuning still lacks robustness to textual adversarial attacks. However, most currently developed defense techniques necessitate auxiliary model update and storage, which inevitably hamper the modularity and low storage of prefix-tuning. In this work, we propose a robust prefix-tuning framework that preserves the efficiency and modularity of prefix-tuning. The core idea of our framework is leveraging the layerwise activations of the language model by correctly-classified training data as the standard for additional prefix finetuning. During the test phase, an extra batch-level prefix is tuned for each batch and added to the original prefix for robustness enhancement. Extensive experiments on three text classification benchmarks show that our framework substantially improves robustness over several strong baselines against five textual attacks of different types while maintaining comparable accuracy on clean texts. We also interpret our robust prefix-tuning framework from the optimal control perspective and pose several directions for future research.

The big picture: Psychedelic drugs have long been touted as possible treatments for mental-health disorders like depression and PTSD. But very little is really known about what these substances actually do to our brains. Understanding how they work could help unlock their potential. A new methodology: Some scientists are using AI to figure it out. A team at McGill University in Montreal used natural language processing to study written "trip reports" of users' experiences with a range of drugs.

The Genius 4.0 Home Intelligence update adds Siri Shortcut Integration to the iRobot Home app, allowing iOS users to connect their devices to Apple's voice assistant. Similar to Google Assistant and Alexa users, they can set up their custom phrases or simply say "Hey Siri, ask Roomba to clean everywhere" to start the vacuum. Genius 4.0 also gives users the capability to create customizable smart maps for the Roomba i3 and i3 models, which they can access if they want their devices to clean specific rooms in the house. They can also create custom cleaning routines based on their schedules, automatons and the rooms they want to send the vacuum to. These particular features are now available in the Americas and will make their way to customers in Europe, Middle East and Africa by the end of the third quarter.