The Edge computing paradigm has gained prominence in both academic and industry circles in recent years. By implementing edge computing facilities and services in robotics, it becomes a key enabler in the deployment of artificial intelligence applications to robots. Time-sensitive robotics applications benefit from the reduced latency, mobility, and location awareness provided by the edge computing paradigm, which enables real-time data processing and intelligence at the network's edge. While the advantages of integrating edge computing into robotics are numerous, there has been no recent survey that comprehensively examines these benefits. This paper aims to bridge that gap by highlighting important work in the domain of edge robotics, examining recent advancements, and offering deeper insight into the challenges and motivations behind both current and emerging solutions. In particular, this article provides a comprehensive evaluation of recent developments in edge robotics, with an emphasis on fundamental applications, providing in-depth analysis of the key motivations, challenges, and future directions in this rapidly evolving domain. It also explores the importance of edge computing in real-world robotics scenarios where rapid response times are critical. Finally, the paper outlines various open research challenges in the field of edge robotics.

We propose a stochastic framework for wildfire spread prediction using deep generative diffusion models with ensemble sampling. In contrast to traditional deterministic approaches that struggle to capture the inherent uncertainty and variability of wildfire dynamics, our method generates probabilistic forecasts by sampling multiple plausible future scenarios conditioned on the same initial state. As a proof-of-concept, the model is trained on synthetic wildfire data generated by a probabilistic cellular automata-based simulator, which integrates realistic environmental features such as canopy cover, vegetation density, and terrain slope, and is grounded in historical fire events including the Chimney and Ferguson fires. To assess predictive performance and uncertainty modelling, we compare two surrogate models with identical network architecture: one trained via conventional supervised regression, and the other using a conditional diffusion framework with ensemble sampling. In the diffusion-based emulator, multiple inference passes are performed for the same input state by resampling the initial latent variable, allowing the model to capture a distribution of possible outcomes.

Semantic data representations are critical in artificial intelligence, significantly enhancing model performance in tasks like transfer and zero-shot learning (Lake et al., 2017). Central to this effort is to disentangle latent representations in generative models--representations where each latent dimension corresponds to an independent underlying factor of variation in the data. Disentanglement is achieved by leveraging statistical properties of the latent space and the dataset, enabling models where changes in one latent dimension affect only its corresponding factor without impacting others. This not only improves model interpretability but also enhances robustness against adversarial attacks (Yang et al., 2021). For a comprehensive review of disentanglement and its statistical underpinnings, see Wang et al. (2023). Datasets encountered in scientific research are often heterogeneous in modalities, fidelities, and accuracy where a particular entity or a state may be simultaneously associated with multiple images, graphs, vectors, scalar parameters, or labels with various associated measurement uncertainties.

Forecasting complex time series is an important yet challenging problem that involves various industrial applications. Recently, masked time-series modeling has been proposed to effectively model temporal dependencies for forecasting by reconstructing masked segments from unmasked ones. However, since the semantic information in time series is involved in intricate temporal variations generated by multiple time series components, simply masking a raw time series ignores the inherent semantic structure, which may cause MTM to learn spurious temporal patterns present in the raw data. To capture distinct temporal semantics, we show that masked modeling techniques should address entangled patterns through a decomposition approach. Specifically, we propose ST-MTM, a masked time-series modeling framework with seasonal-trend decomposition, which includes a novel masking method for the seasonal-trend components that incorporates different temporal variations from each component. ST-MTM uses a period masking strategy for seasonal components to produce multiple masked seasonal series based on inherent multi-periodicity and a sub-series masking strategy for trend components to mask temporal regions that share similar variations. The proposed masking method presents an effective pre-training task for learning intricate temporal variations and dependencies. Additionally, ST-MTM introduces a contrastive learning task to support masked modeling by enhancing contextual consistency among multiple masked seasonal representations. Experimental results show that our proposed ST-MTM achieves consistently superior forecasting performance compared to existing masked modeling, contrastive learning, and supervised forecasting methods.

We study the applicability of GNNs to the problem of wind energy forecasting. We find that certain architectures achieve performance comparable to our best CNN-based benchmark. The study is conducted on three wind power facilities using five years of historical data. Numerical Weather Prediction (NWP) variables were used as predictors, and models were evaluated on a 24 to 36 hour ahead test horizon.

Artificial intelligence (AI) is fueling exponential electricity demand growth, threatening grid reliability, raising prices for communities paying for new energy infrastructure, and stunting AI innovation as data centers wait for interconnection to constrained grids. This paper presents the first field demonstration, in collaboration with major corporate partners, of a software-only approach--Emerald Conductor--that transforms AI data centers into flexible grid resources that can efficiently and immediately harness existing power systems without massive infrastructure buildout. Conducted at a 256-GPU cluster running representative AI workloads within a commercial, hyperscale cloud data center in Phoenix, Arizona, the trial achieved a 25% reduction in cluster power usage for three hours during peak grid events while maintaining AI quality of service (QoS) guarantees. By orchestrating AI workloads based on real-time grid signals without hardware modifications or energy storage, this platform reimagines data centers as grid-interactive assets that enhance grid reliability, advance affordability, and accelerate AI's development.

Water consumption is an increasingly critical dimension of computing sustainability, especially as AI workloads rapidly scale. However, current water impact assessment often overlooks where and when water stress is more severe. To fill in this gap, we present SCARF, the first general framework that evaluates water impact of computing by factoring in both spatial and temporal variations in water stress. SCARF calculates an Adjusted Water Impact (AWI) metric that considers both consumption volume and local water stress over time. Through three case studies on LLM serving, datacenters, and semiconductor fabrication plants, we show the hidden opportunities for reducing water impact by optimizing location and time choices, paving the way for water-sustainable computing. The code is available at https://github.com/jojacola/SCARF.

Many of the observations we make are biased by our decisions. For instance, the demand of items is impacted by the prices set, and online checkout choices are influenced by the assortments presented. The challenge in decision-making under this setting is the lack of counterfactual information, and the need to learn it instead. We introduce an end-to-end method under endogenous uncertainty to train ML models to be aware of their downstream, enabling their effective use in the decision-making stage. We further introduce a robust optimization variant that accounts for uncertainty in ML models -- specifically by constructing uncertainty sets over the space of ML models and optimizing actions to protect against worst-case predictions. We prove guarantees that this robust approach can capture near-optimal decisions with high probability as a function of data. Besides this, we also introduce a new class of two-stage stochastic optimization problems to the end-to-end learning framework that can now be addressed through our framework. Here, the first stage is an information-gathering problem to decide which random variable to poll and gain information about before making a second-stage decision based off of it. We present several computational experiments for pricing and inventory assortment/recommendation problems. We compare against existing methods in online learning/bandits/offline reinforcement learning and show our approach has consistent improved performance over these. Just as in the endogenous setting, the model's prediction also depends on the first-stage decision made. While this decision does not affect the random variable in this setting, it does affect the correct point forecast that should be made.

Recent advances in word embeddings and language models use large-scale, unlabelled data and self-supervised learning to boost NLP performance. Multilingual models, often trained on web-sourced data like Wikipedia, face challenges: few low-resource languages are included, their data is often noisy, and lack of labeled datasets makes it hard to evaluate performance outside high-resource languages like English. In this dissertation, we focus on languages spoken in Sub-Saharan Africa where all the indigenous languages in this region can be regarded as low-resourced in terms of the availability of labelled data for NLP tasks and unlabelled data found on the web. We analyse the noise in the publicly available corpora, and curate a high-quality corpus, demonstrating that the quality of semantic representations learned in word embeddings does not only depend on the amount of data but on the quality of pre-training data. We demonstrate empirically the limitations of word embeddings, and the opportunities the multilingual pre-trained language model (PLM) offers especially for languages unseen during pre-training and low-resource scenarios. We further study how to adapt and specialize multilingual PLMs to unseen African languages using a small amount of monolingual texts. To address the under-representation of the African languages in NLP research, we developed large scale human-annotated labelled datasets for 21 African languages in two impactful NLP tasks: named entity recognition and machine translation. We conduct an extensive empirical evaluation using state-of-the-art methods across supervised, weakly-supervised, and transfer learning settings.

Accurate traffic prediction is essential for Intelligent Transportation Systems (ITS), yet current methods struggle with the inherent complexity and non-linearity of traffic dynamics, making it difficult to integrate spatial and temporal characteristics. Furthermore, existing approaches use static techniques to address non-stationary and anomalous historical data, which limits adaptability and undermines data smoothing. To overcome these challenges, we propose the Graph Fusion Enhanced Network (GFEN), an innovative framework for network-level traffic speed prediction. GFEN introduces a novel topological spatiotemporal graph fusion technique that meticulously extracts and merges spatial and temporal correlations from both data distribution and network topology using trainable methods, enabling the modeling of multi-scale spatiotemporal features. Additionally, GFEN employs a hybrid methodology combining a k-th order difference-based mathematical framework with an attention-based deep learning structure to adaptively smooth historical observations and dynamically mitigate data anomalies and non-stationarity. Extensive experiments demonstrate that GFEN surpasses state-of-the-art methods by approximately 6.3% in prediction accuracy and exhibits convergence rates nearly twice as fast as recent hybrid models, confirming its superior performance and potential to significantly enhance traffic prediction system efficiency.