Ensemble datasets are ever more prevalent in various scientific domains. In climate science, ensemble datasets are used to capture variability in projections under plausible future conditions including greenhouse and aerosol emissions. Each ensemble model run produces projections that are fundamentally similar yet meaningfully distinct. Understanding this variability among ensemble model runs and analyzing its magnitude and patterns is a vital task for climate scientists. In this paper, we present ClimateSOM, a visual analysis workflow that leverages a self-organizing map (SOM) and Large Language Models (LLMs) to support interactive exploration and interpretation of climate ensemble datasets. The workflow abstracts climate ensemble model runs - spatiotemporal time series - into a distribution over a 2D space that captures the variability among the ensemble model runs using a SOM. LLMs are integrated to assist in sensemaking of this SOM-defined 2D space, the basis for the visual analysis tasks. In all, ClimateSOM enables users to explore the variability among ensemble model runs, identify patterns, compare and cluster the ensemble model runs. To demonstrate the utility of ClimateSOM, we apply the workflow to an ensemble dataset of precipitation projections over California and the Northwestern United States. Furthermore, we conduct a short evaluation of our LLM integration, and conduct an expert review of the visual workflow and the insights from the case studies with six domain experts to evaluate our approach and its utility.

Deep Reinforcement Learning maybe be the most accurate model of Machine Learning to an actual human mind. What is so special about them? I will answer that for you, when deep RL combined with attention mechanism goes in some other league, something powerful enough to understand what is happening and what is the purpose of the machine (Maybe something similar to ultron from Avengers). A new term is tossed when these two are combined known as DRARL (Deep Residual Attention Reinforcement Learning). DRARL models have the ability to understand a question and then answer it, another ability it has, is to understand patterns from a given data.

MLflow is a great open source tool that allows you to track your model runs, including model parameters, metrics, results, data used, and your code. MLflow also has many other capabilities such as deploying models, packaging your code for reproducibility, and storing your models. For this introduction, we will focus on tracking our model runs with MLflow. Tracking your model runs can be very useful in order to see the differences model performance based on the parameters and data used for each run. In this tutorial, we will integrate MLflow into our machine learning workflow.

As the pandemic has made big changes across our world, we can't always rely on historical data that we used to train and build our first model versions. We all know -- or we should realize by now -- that these first versions will break somehow. It is just a matter of time. In our first article from last December, we discussed why you need model monitoring on your AI/ML models. Let's broaden that discussion, and our vision, by considering a holistic framework to maintain our models.

This Overview, Design Concepts, and Details Protocol (ODD) provides a detailed description of an agent-based model (ABM) that was developed to simulate hospitalizations during the COVID-19 pandemic. Using the descriptions of submodels, provided parameters, and the links to data sources, modelers will be able to replicate the creation and results of this model.

Computer models play a key role in many scientific and engineering problems. One major source of uncertainty in computer model experiment is input parameter uncertainty. Computer model calibration is a formal statistical procedure to infer input parameters by combining information from model runs and observational data. The existing standard calibration framework suffers from inferential issues when the model output and observational data are high-dimensional dependent data such as large time series due to the difficulty in building an emulator and the non-identifiability between effects from input parameters and data-model discrepancy. To overcome these challenges we propose a new calibration framework based on a deep neural network (DNN) with long-short term memory layers that directly emulates the inverse relationship between the model output and input parameters. Adopting the 'learning with noise' idea we train our DNN model to filter out the effects from data model discrepancy on input parameter inference. We also formulate a new way to construct interval predictions for DNN using quantile regression to quantify the uncertainty in input parameter estimates. Through a simulation study and real data application with WRF-hydro model we show that our approach can yield accurate point estimates and well calibrated interval estimates for input parameters.

Artificial intelligence vision silicon company Ambarella is partnering with Amazon Web Services to allow AWS customers to use the tech giant's services to train machine learning models and run them on devices equipped with Ambarella's CVflow AI vision chip. According to Ambarella, developers previously had to manually optimize machine learning models for devices based on the company's AI vision system on chip (SOC), a step that could add delays and errors to the app development process. In an announcement, the companies said they collaborated to simplify the process by integrating the Ambarella toolchain with the Amazon SageMaker Neo cloud service. Now, developers can bring trained models to Amazon SageMaker Neo and automatically optimize the model for Ambarella's CVflow-powered SoCs, the companies said. Using MXNet, TensorFlow, PyTorch or XGBoost, customers can train the model using Amazon SageMaker in the cloud or their local machine.

In my experience, learning to anything useful in computer science has fallen at the strange intersection of theory and practice. It's pretty easy to ignore the amount of depth that the lies under some of the things we code. Machine learning takes that to an extreme, and everyone wants to be a Machine Learning Engineer these days. Elements of Statistical Learning is a fantastic book. If you can get through it you'll know quite a bit, but it doesn't mean much if you're unable to put any of into practice.

Surrogate strategies are used widely for uncertainty quantification of groundwater models in order to improve computational efficiency. However, their application to dynamic multiphase flow problems is hindered by the curse of dimensionality, the saturation discontinuity due to capillarity effects, and the time-dependence of the multi-output responses. In this paper, we propose a deep convolutional encoder-decoder neural network methodology to tackle these issues. The surrogate modeling task is transformed to an image-to-image regression strategy. This approach extracts high-level coarse features from the high-dimensional input permeability images using an encoder, and then refines the coarse features to provide the output pressure/saturation images through a decoder. A training strategy combining a regression loss and a segmentation loss is proposed in order to better approximate the discontinuous saturation field. To characterize the high-dimensional time-dependent outputs of the dynamic system, time is treated as an additional input to the network that is trained using pairs of input realizations and of the corresponding system outputs at a limited number of time instances. The proposed method is evaluated using a geological carbon storage process-based multiphase flow model with a 2500-dimensional stochastic permeability field. With a relatively small number of training data, the surrogate model is capable of accurately characterizing the spatio-temporal evolution of the pressure and discontinuous CO2 saturation fields and can be used efficiently to compute the statistics of the system responses.

Direct numerical simulation of Stokes flow through an impermeable, rigid body matrix by finite elements requires meshes fine enough to resolve the pore-size scale and is thus a computationally expensive task. The cost is significantly amplified when randomness in the pore microstructure is present and therefore multiple simulations need to be carried out. It is well known that in the limit of scale-separation, Stokes flow can be accurately approximated by Darcy's law with an effective diffusivity field depending on viscosity and the pore-matrix topology. We propose a fully probabilistic, Darcy-type, reduced-order model which, based on only a few tens of full-order Stokes model runs, is capable of learning a map from the fine-scale topology to the effective diffusivity and is maximally predictive of the fine-scale response. The reduced-order model learned can significantly accelerate uncertainty quantification tasks as well as provide quantitative confidence metrics of the predictive estimates produced.