Inflation is causing stress and worry around the globe. Residents of Turkey, Argentina, Venezuela and dozens of other countries have seen prices jump by hundreds of percent in recent years. And this is becoming a larger problem in developed countries as well. According to experts at Goldman Sachs, U.K. inflation could hit as high as 22% in the coming period. The European Union also faces sharp price rises.

Quechua language and Quechua knowledge gather millions of people around the world, especially in several countries in South America. Unfortunately, there are only a few resources available to Quechua communities, and they are mainly stored in PDF format. In this paper, the Quechua Knowledge Graph is envisioned and generated as an effort to get Quechua closer to the Quechua communities, researchers, and technology developers. Currently, there are 553636 triples stored in the Quechua Knowledge Graph, which is accessible on the Web, retrievable by machines, and curated by users. To showcase the deployment of the Quechua Knowledge Graph, use cases and future work are described.

The temporal and spatial resolution of rainfall data is crucial for environmental modeling studies in which its variability in space and time is considered as a primary factor. Rainfall products from different remote sensing instruments (e.g., radar, satellite) have different space-time resolutions because of the differences in their sensing capabilities and post-processing methods. In this study, we developed a deep learning approach that augments rainfall data with increased time resolutions to complement relatively lower resolution products. We propose a neural network architecture based on Convolutional Neural Networks (CNNs) to improve the temporal resolution of radar-based rainfall products and compare the proposed model with an optical flow-based interpolation method and CNN-baseline model. The methodology presented in this study could be used for enhancing rainfall maps with better temporal resolution and imputation of missing frames in sequences of 2D rainfall maps to support hydrological and flood forecasting studies.

In lifelong learning, an agent learns throughout its entire life without resets, in a constantly changing environment, as we humans do. Consequently, lifelong learning comes with a plethora of research problems such as continual domain shifts, which result in non-stationary rewards and environment dynamics. These non-stationarities are difficult to detect and cope with due to their continuous nature. Therefore, exploration strategies and learning methods are required that are capable of tracking the steady domain shifts, and adapting to them. We propose Reactive Exploration to track and react to continual domain shifts in lifelong reinforcement learning, and to update the policy correspondingly. To this end, we conduct experiments in order to investigate different exploration strategies. We empirically show that representatives of the policy-gradient family are better suited for lifelong learning, as they adapt more quickly to distribution shifts than Q-learning. Thereby, policy-gradient methods profit the most from Reactive Exploration and show good results in lifelong learning with continual domain shifts. Our code is available at: https://github.com/ml-jku/reactive-exploration.

Accented text-to-speech (TTS) synthesis seeks to generate speech with an accent (L2) as a variant of the standard version (L1). Accented TTS synthesis is challenging as L2 is different from L1 in both in terms of phonetic rendering and prosody pattern. Furthermore, there is no easy solution to the control of the accent intensity in an utterance. In this work, we propose a neural TTS architecture, that allows us to control the accent and its intensity during inference. This is achieved through three novel mechanisms, 1) an accent variance adaptor to model the complex accent variance with three prosody controlling factors, namely pitch, energy and duration; 2) an accent intensity modeling strategy to quantify the accent intensity; 3) a consistency constraint module to encourage the TTS system to render the expected accent intensity at a fine level. Experiments show that the proposed system attains superior performance to the baseline models in terms of accent rendering and intensity control. To our best knowledge, this is the first study of accented TTS synthesis with explicit intensity control.

A startup backed by a billionaire Brazilian family is partnering with e-commerce platform Farmers Business Network to offer robots that spray fertilizer and pesticides to US farmers. Solinftec's robots, which run on solar panels, are autonomous and were designed to apply fertilizer and weed killer only where needed. The company said the technology, already in use in Brazil, can reduce product use by as much as 70%. As a result of the deal, both companies will work with farmers in the US to commercialize the robot and to develop new methods for agrochemicals to be used with robotics technology.

My father and my brothers are amateur musicians who play various instruments; thus, music runs in my family. I was raised in a household where music listening is a tradition and where kids are encouraged to sing, play, and listen to music early in our days in Brazil. My earliest memories are of music; they are the things that moved me the most and that life taught me to value the most. I put together some rock bands when I was a teenager, and even now, there is never a shortage of musical instruments at my house, both analog and now digital. In addition to my musical education, adult life led me along other routes, including those in computing and artificial intelligence: a really rich experience that allowed me to learn new languages and develop technical skills.

Text classification has become widely used in various natural language processing applications like sentiment analysis. Current applications often use large transformer-based language models to classify input texts. However, there is a lack of systematic study on how much private information can be inverted when publishing models. In this paper, we formulate \emph{Text Revealer} -- the first model inversion attack for text reconstruction against text classification with transformers. Our attacks faithfully reconstruct private texts included in training data with access to the target model. We leverage an external dataset and GPT-2 to generate the target domain-like fluent text, and then perturb its hidden state optimally with the feedback from the target model. Our extensive experiments demonstrate that our attacks are effective for datasets with different text lengths and can reconstruct private texts with accuracy.

This work explores the physics-driven machine learning technique Operator Inference (OpInf) for predicting the state of chaotic dynamical systems. OpInf provides a non-intrusive approach to infer approximations of polynomial operators in reduced space without having access to the full order operators appearing in discretized models. Datasets for the physics systems are generated using conventional numerical solvers and then projected to a low-dimensional space via Principal Component Analysis (PCA). In latent space, a least-squares problem is set to fit a quadratic polynomial operator, which is subsequently employed in a time-integration scheme in order to produce extrapolations in the same space. Once solved, the inverse PCA operation is applied to reconstruct the extrapolations in the original space. The quality of the OpInf predictions is assessed via the Normalized Root Mean Squared Error (NRMSE) metric from which the Valid Prediction Time (VPT) is computed. Numerical experiments considering the chaotic systems Lorenz 96 and the Kuramoto-Sivashinsky equation show promising forecasting capabilities of the OpInf reduced order models with VPT ranges that outperform state-of-the-art machine learning methods such as backpropagation and reservoir computing recurrent neural networks [1], as well as Markov neural operators [2].

During recent years transformers architectures have been growing in popularity. Modulated Detection Transformer (MDETR) is an end-to-end multi-modal understanding model that performs tasks such as phase grounding, referring expression comprehension, referring expression segmentation, and visual question answering. One remarkable aspect of the model is the capacity to infer over classes that it was not previously trained for. In this work we explore the use of MDETR in a new task, action detection, without any previous training. We obtain quantitative results using the Atomic Visual Actions dataset. Although the model does not report the best performance in the task, we believe that it is an interesting finding. We show that it is possible to use a multi-modal model to tackle a task that it was not designed for. Finally, we believe that this line of research may lead into the generalization of MDETR in additional downstream tasks.