In recent years, the development of accurate deep keyword spotting (KWS) models has resulted in KWS technology being embedded in a number of technologies such as voice assistants. Many of these models rely on large amounts of labelled data to achieve good performance. As a result, their use is restricted to applications for which a large labelled speech data set can be obtained. Self-supervised learning seeks to mitigate the need for large labelled data sets by leveraging unlabelled data, which is easier to obtain in large amounts. However, most self-supervised methods have only been investigated for very large models, whereas KWS models are desired to be small. In this paper, we investigate the use of self-supervised pretraining for the smaller KWS models in a label-deficient scenario. We pretrain the Keyword Transformer model using the self-supervised framework Data2Vec and carry out experiments on a label-deficient setup of the Google Speech Commands data set. It is found that the pretrained models greatly outperform the models without pretraining, showing that Data2Vec pretraining can increase the performance of KWS models in label-deficient scenarios. The source code is made publicly available.

The occurrence of vacuum arcs or radio frequency (rf) breakdowns is one of the most prevalent factors limiting the high-gradient performance of normal conducting rf cavities in particle accelerators. In this paper, we search for the existence of previously unrecognized features related to the incidence of rf breakdowns by applying a machine learning strategy to high-gradient cavity data from CERN's test stand for the Compact Linear Collider (CLIC). By interpreting the parameters of the learned models with explainable artificial intelligence (AI), we reverse-engineer physical properties for deriving fast, reliable, and simple rule-based models. Based on 6 months of historical data and dedicated experiments, our models show fractions of data with a high influence on the occurrence of breakdowns. Specifically, it is shown that the field emitted current following an initial breakdown is closely related to the probability of another breakdown occurring shortly thereafter. Results also indicate that the cavity pressure should be monitored with increased temporal resolution in future experiments, to further explore the vacuum activity associated with breakdowns.

In this paper, we present an evolved version of the Situational Graphs, which jointly models in a single optimizable factor graph, a SLAM graph, as a set of robot keyframes, containing its associated measurements and robot poses, and a 3D scene graph, as a high-level representation of the environment that encodes its different geometric elements with semantic attributes and the relational information between those elements. Our proposed S-Graphs+ is a novel four-layered factor graph that includes: (1) a keyframes layer with robot pose estimates, (2) a walls layer representing wall surfaces, (3) a rooms layer encompassing sets of wall planes, and (4) a floors layer gathering the rooms within a given floor level. The above graph is optimized in real-time to obtain a robust and accurate estimate of the robot's pose and its map, simultaneously constructing and leveraging the high-level information of the environment. To extract such high-level information, we present novel room and floor segmentation algorithms utilizing the mapped wall planes and free-space clusters. We tested S-Graphs+ on multiple datasets including, simulations of distinct indoor environments, on real datasets captured over several construction sites and office environments, and on a real public dataset of indoor office environments. S-Graphs+ outperforms relevant baselines in the majority of the datasets while extending the robot situational awareness by a four-layered scene model. Moreover, we make the algorithm available as a docker file.

Machine learning methods like neural networks are extremely successful and popular in a variety of applications, however, they come at substantial computational costs, accompanied by high energy demands. In contrast, hardware capabilities are limited and there is evidence that technology scaling is stuttering, therefore, new approaches to meet the performance demands of increasingly complex model architectures are required. As an unsafe optimization, noisy computations are more energy efficient, and given a fixed power budget also more time efficient. However, any kind of unsafe optimization requires counter measures to ensure functionally correct results. This work considers noisy computations in an abstract form, and gears to understand the implications of such noise on the accuracy of neural-network-based classifiers as an exemplary workload. We propose a methodology called "Walking Noise" that allows to assess the robustness of different layers of deep architectures by means of a so-called "midpoint noise level" metric. We then investigate the implications of additive and multiplicative noise for different classification tasks and model architectures, with and without batch normalization. While noisy training significantly increases robustness for both noise types, we observe a clear trend to increase weights and thus increase the signal-to-noise ratio for additive noise injection. For the multiplicative case, we find that some networks, with suitably simple tasks, automatically learn an internal binary representation, hence becoming extremely robust. Overall this work proposes a method to measure the layer-specific robustness and shares first insights on how networks learn to compensate injected noise, and thus, contributes to understand robustness against noisy computations.

End-to-End speech-to-speech translation (S2ST) is generally evaluated with text-based metrics. This means that generated speech has to be automatically transcribed, making the evaluation dependent on the availability and quality of automatic speech recognition (ASR) systems. In this paper, we propose a text-free evaluation metric for end-to-end S2ST, named BLASER, to avoid the dependency on ASR systems. BLASER leverages a multilingual multimodal encoder to directly encode the speech segments for source input, translation output and reference into a shared embedding space and computes a score of the translation quality that can be used as a proxy to human evaluation. To evaluate our approach, we construct training and evaluation sets from more than 40k human annotations covering seven language directions. The best results of BLASER are achieved by training with supervision from human rating scores. We show that when evaluated at the sentence level, BLASER correlates significantly better with human judgment compared to ASR-dependent metrics including ASR-SENTBLEU in all translation directions and ASR-COMET in five of them. Our analysis shows combining speech and text as inputs to BLASER does not increase the correlation with human scores, but best correlations are achieved when using speech, which motivates the goal of our research. Moreover, we show that using ASR for references is detrimental for text-based metrics.

Compressing neural network architectures is important to allow the deployment of models to embedded or mobile devices, and pruning and quantization are the major approaches to compress neural networks nowadays. Both methods benefit when compression parameters are selected specifically for each layer. Finding good combinations of compression parameters, so-called compression policies, is hard as the problem spans an exponentially large search space. Effective compression policies consider the influence of the specific hardware architecture on the used compression methods. We propose an algorithmic framework called Galen to search such policies using reinforcement learning utilizing pruning and quantization, thus providing automatic compression for neural networks. Contrary to other approaches we use inference latency measured on the target hardware device as an optimization goal. With that, the framework supports the compression of models specific to a given hardware target. We validate our approach using three different reinforcement learning agents for pruning, quantization and joint pruning and quantization. Besides proving the functionality of our approach we were able to compress a ResNet18 for CIFAR-10, on an embedded ARM processor, to 20% of the original inference latency without significant loss of accuracy. Moreover, we can demonstrate that a joint search and compression using pruning and quantization is superior to an individual search for policies using a single compression method.

Efficient localization plays a vital role in many modern applications of Unmanned Ground Vehicles (UGV) and Unmanned aerial vehicles (UAVs), which would contribute to improved control, safety, power economy, etc. The ubiquitous 5G NR (New Radio) cellular network will provide new opportunities for enhancing localization of UAVs and UGVs. In this paper, we review the radio frequency (RF) based approaches for localization. We review the RF features that can be utilized for localization and investigate the current methods suitable for Unmanned vehicles under two general categories: range-based and fingerprinting. The existing state-of-the-art literature on RF-based localization for both UAVs and UGVs is examined, and the envisioned 5G NR for localization enhancement, and the future research direction are explored.

Understanding deep learning model behavior is critical to accepting machine learning-based decision support systems in the medical community. Previous research has shown that jointly using clinical notes with electronic health record (EHR) data improved predictive performance for patient monitoring in the intensive care unit (ICU). In this work, we explore the underlying reasons for these improvements. While relying on a basic attention-based model to allow for interpretability, we first confirm that performance significantly improves over state-of-the-art EHR data models when combining EHR data and clinical notes. We then provide an analysis showing improvements arise almost exclusively from a subset of notes containing broader context on patient state rather than clinician notes. We believe such findings highlight deep learning models for EHR data to be more limited by partially-descriptive data than by modeling choice, motivating a more data-centric approach in the field.

With the rise of AI in recent years and the increase in complexity of the models, the growing demand in computational resources is starting to pose a significant challenge. The need for higher compute power is being met with increasingly more potent accelerators and the use of large compute clusters. However, the gain in prediction accuracy from large models trained on distributed and accelerated systems comes at the price of a substantial increase in energy demand, and researchers have started questioning the environmental friendliness of such AI methods at scale. Consequently, energy efficiency plays an important role for AI model developers and infrastructure operators alike. The energy consumption of AI workloads depends on the model implementation and the utilized hardware. Therefore, accurate measurements of the power draw of AI workflows on different types of compute nodes is key to algorithmic improvements and the design of future compute clusters and hardware. To this end, we present measurements of the energy consumption of two typical applications of deep learning models on different types of compute nodes. Our results indicate that 1. deriving energy consumption directly from runtime is not accurate, but the consumption of the compute node needs to be considered regarding its composition; 2. neglecting accelerator hardware on mixed nodes results in overproportional inefficiency regarding energy consumption; 3. energy consumption of model training and inference should be considered separately - while training on GPUs outperforms all other node types regarding both runtime and energy consumption, inference on CPU nodes can be comparably efficient. One advantage of our approach is that the information on energy consumption is available to all users of the supercomputer, enabling an easy transfer to other workloads alongside a raise in user-awareness of energy consumption.

This work addresses cross-view camera pose estimation, i.e., determining the 3-DoF camera pose of a given ground-level image w.r.t. an aerial image of the local area. We propose SliceMatch, which consists of ground and aerial feature extractors, feature aggregators, and a pose predictor. The feature extractors extract dense features from the ground and aerial images. Given a set of candidate camera poses, the feature aggregators construct a single ground descriptor and a set of rotational equivariant pose-dependent aerial descriptors. Notably, our novel aerial feature aggregator has a cross-view attention module for ground-view guided aerial feature selection, and utilizes the geometric projection of the ground camera's viewing frustum on the aerial image to pool features. The efficient construction of aerial descriptors is achieved by using precomputed masks and by re-assembling the aerial descriptors for rotated poses. SliceMatch is trained using contrastive learning and pose estimation is formulated as a similarity comparison between the ground descriptor and the aerial descriptors. SliceMatch outperforms the state-of-the-art by 19% and 62% in median localization error on the VIGOR and KITTI datasets, with 3x FPS of the fastest baseline.