Recent video+language datasets cover domains where the interaction is highly structured, such as instructional videos, or where the interaction is scripted, such as TV shows. Both of these properties can lead to spurious cues to be exploited by models rather than learning to ground language. In this paper, we present GrOunded footbAlL commentaries (GOAL), a novel dataset of football (or `soccer') highlights videos with transcribed live commentaries in English. As the course of a game is unpredictable, so are commentaries, which makes them a unique resource to investigate dynamic language grounding. We also provide state-of-the-art baselines for the following tasks: frame reordering, moment retrieval, live commentary retrieval and play-by-play live commentary generation. Results show that SOTA models perform reasonably well in most tasks. We discuss the implications of these results and suggest new tasks for which GOAL can be used. Our codebase is available at: https://gitlab.com/grounded-sport-convai/goal-baselines.

In this work a novel recommender system (RS) for Tourism is presented. The RS is context aware as is now the rule in the state-of-the-art for recommender systems and works on top of a tourism ontology which is used to group the different items being offered. The presented RS mixes different types of recommenders creating an ensemble which changes on the basis of the RS's maturity. Starting from simple content-based recommendations and iteratively adding popularity, demographic and collaborative filtering methods as rating density and user cardinality increases. The result is a RS that mutates during its lifetime and uses a tourism ontology and natural language processing (NLP) to correctly bin the items to specific item categories and meta categories in the ontology. This item classification facilitates the association between user preferences and items, as well as allowing to better classify and group the items being offered, which in turn is particularly useful for context-aware filtering.

Detecting anomalous data within time series is a very relevant task in pattern recognition and machine learning, with many possible applications that range from disease prevention in medicine, e.g., detecting early alterations of the health status before it can clearly be defined as "illness" up to monitoring industrial plants. Regarding this latter application, detecting anomalies in an industrial plant's status firstly prevents serious damages that would require a long interruption of the production process. Secondly, it permits optimal scheduling of maintenance interventions by limiting them to urgent situations. At the same time, they typically follow a fixed prudential schedule according to which components are substituted well before the end of their expected lifetime. This paper describes a case study regarding the monitoring of the status of Laser-guided Vehicles (LGVs) batteries, on which we worked as our contribution to project SUPER (Supercomputing Unified Platform, Emilia Romagna) aimed at establishing and demonstrating a regional High-Performance Computing platform that is going to represent the main Italian supercomputing environment for both computing power and data volume.

Neural style transfer is a deep learning technique that produces an unprecedentedly rich style transfer from a style image to a content image and is particularly impressive when it comes to transferring style from a painting to an image. It was originally achieved by solving an optimization problem to match the global style statistics of the style image while preserving the local geometric features of the content image. The two main drawbacks of this original approach is that it is computationally expensive and that the resolution of the output images is limited by high GPU memory requirements. Many solutions have been proposed to both accelerate neural style transfer and increase its resolution, but they all compromise the quality of the produced images. Indeed, transferring the style of a painting is a complex task involving features at different scales, from the color palette and compositional style to the fine brushstrokes and texture of the canvas. This paper provides a solution to solve the original global optimization for ultra-high resolution images, enabling multiscale style transfer at unprecedented image sizes. This is achieved by spatially localizing the computation of each forward and backward passes through the VGG network. Extensive qualitative and quantitative comparisons show that our method produces a style transfer of unmatched quality for such high resolution painting styles.

Artificial intelligence (AI) in the form of deep learning bears promise for drug discovery and chemical biology, $\textit{e.g.}$, to predict protein structure and molecular bioactivity, plan organic synthesis, and design molecules $\textit{de novo}$. While most of the deep learning efforts in drug discovery have focused on ligand-based approaches, structure-based drug discovery has the potential to tackle unsolved challenges, such as affinity prediction for unexplored protein targets, binding-mechanism elucidation, and the rationalization of related chemical kinetic properties. Advances in deep learning methodologies and the availability of accurate predictions for protein tertiary structure advocate for a $\textit{renaissance}$ in structure-based approaches for drug discovery guided by AI. This review summarizes the most prominent algorithmic concepts in structure-based deep learning for drug discovery, and forecasts opportunities, applications, and challenges ahead.

This work presents a set of neural network (NN) models specifically designed for accurate and efficient fluid dynamics forecasting. In this work, we show how neural networks training can be improved by reducing data complexity through a modal decomposition technique called higher order dynamic mode decomposition (HODMD), which identifies the main structures inside flow dynamics and reconstructs the original flow using only these main structures. This reconstruction has the same number of samples and spatial dimension as the original flow, but with a less complex dynamics and preserving its main features. We also show the low computational cost required by the proposed NN models, both in their training and inference phases. The core idea of this work is to test the limits of applicability of deep learning models to data forecasting in complex fluid dynamics problems. Generalization capabilities of the models are demonstrated by using the same neural network architectures to forecast the future dynamics of four different multi-phase flows. Data sets used to train and test these deep learning models come from Direct Numerical Simulations (DNS) of these flows.

Network models are an essential block of modern networks. For example, they are widely used in network planning and optimization. However, as networks increase in scale and complexity, some models present limitations, such as the assumption of markovian traffic in queuing theory models, or the high computational cost of network simulators. Recent advances in machine learning, such as Graph Neural Networks (GNN), are enabling a new generation of network models that are data-driven and can learn complex non-linear behaviors. In this paper, we present RouteNet-Fermi, a custom GNN model that shares the same goals as queuing theory, while being considerably more accurate in the presence of realistic traffic models. The proposed model predicts accurately the delay, jitter, and loss in networks. We have tested RouteNet-Fermi in networks of increasing size (up to 300 nodes), including samples with mixed traffic profiles -- e.g., with complex non-markovian models -- and arbitrary routing and queue scheduling configurations. Our experimental results show that RouteNet-Fermi achieves similar accuracy as computationally-expensive packet-level simulators and it is able to accurately scale to large networks. For example, the model produces delay estimates with a mean relative error of 6.24% when applied to a test dataset with 1,000 samples, including network topologies one order of magnitude larger than those seen during training.

Data scarcity is one of the main issues with the end-to-end approach for Speech Translation, as compared to the cascaded one. Although most data resources for Speech Translation are originally document-level, they offer a sentence-level view, which can be directly used during training. But this sentence-level view is single and static, potentially limiting the utility of the data. Our proposed data augmentation method SegAugment challenges this idea and aims to increase data availability by providing multiple alternative sentence-level views of a dataset. Our method heavily relies on an Audio Segmentation system to re-segment the speech of each document, after which we obtain the target text with alignment methods. The Audio Segmentation system can be parameterized with different length constraints, thus giving us access to multiple and diverse sentence-level views for each document. Experiments in MuST-C show consistent gains across 8 language pairs, with an average increase of 2.2 BLEU points, and up to 4.7 BLEU for lower-resource scenarios in mTEDx. Additionally, we find that SegAugment is also applicable to purely sentence-level data, as in CoVoST, and that it enables Speech Translation models to completely close the gap between the gold and automatic segmentation at inference time.

We leverage path differentiability and a recent result on nonsmooth implicit differentiation calculus to give sufficient conditions ensuring that the solution to a monotone inclusion problem will be path differentiable, with formulas for computing its generalized gradient. A direct consequence of our result is that these solutions happen to be differentiable almost everywhere. Our approach is fully compatible with automatic differentiation and comes with assumptions which are easy to check, roughly speaking: semialgebraicity and strong monotonicity. We illustrate the scope of our results by considering three fundamental composite problem settings: strongly convex problems, dual solutions to convex minimization problems and primal-dual solutions to min-max problems.

We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging Kinetics-600. (ii) MAGVIT outperforms existing methods in inference time by two orders of magnitude against diffusion models and by 60x against autoregressive models. (iii) A single MAGVIT model supports ten diverse generation tasks and generalizes across videos from different visual domains. The source code and trained models will be released to the public at https://magvit.cs.cmu.edu.