arXiv subjects

Alex Lu

Publications and source records attributed to Alex Lu.

5 recordsLinked to original sources

arXiv2026

Advancing Subseasonal Forecasting with Machine Learning

Decision-makers rely on weather forecasts to plant crops, manage wildfires, allocate water and energy, and prepare for weather extremes. Today, such forecasts enjoy unprecedented accuracy out to two weeks thanks to steady advances in physics-based dynamical models and data-driven artificial intelligence (AI) models. However, model skill drops precipitously at subseasonal timescales (2 - 6 weeks ahead), due to compounding errors, systemic model biases, and the chaotic nature of the atmosphere. To counter this degradation, we introduce probabilistic bias correction (PBC), a machine learning framework that substantially reduces systematic error by learning to correct historical probabilistic forecasts. When applied to the leading dynamical and AI models from the European Centre for Medium-Range Weather Forecasts (ECMWF), PBC doubles the modest subseasonal skill of the AI Forecasting System and improves the skill of the operationally-debiased dynamical model for 91% of pressure, 92% of temperature, and 98% of precipitation targets. We designed PBC for operational deployment, and, in ECMWF's 2025 real-time forecasting competition, its global forecasts placed first for all weather variables and lead times, outperforming the dynamical models from six operational forecasting centers, an international dynamical multi-model ensemble, ECMWF's AI Forecasting System, and the forecasting systems of 34 teams worldwide. These probabilistic skill gains translate into more accurate prediction of extreme events and have the potential to improve agricultural planning, energy management, and disaster preparedness in vulnerable communities.

cs.LG↗

arXiv2023

Domain adaptation using optimal transport for invariant learning using histopathology datasets

Histopathology is critical for the diagnosis of many diseases, including cancer. These protocols typically require pathologists to manually evaluate slides under a microscope, which is time-consuming and subjective, leading to interest in machine learning to automate analysis. However, computational techniques are limited by batch effects, where technical factors like differences in preparation protocol or scanners can alter the appearance of slides, causing models trained on one institution to fail when generalizing to others. Here, we propose a domain adaptation method that improves the generalization of histopathological models to data from unseen institutions, without the need for labels or retraining in these new settings. Our approach introduces an optimal transport (OT) loss, that extends adversarial methods that penalize models if images from different institutions can be distinguished in their representation space. Unlike previous methods, which operate on single samples, our loss accounts for distributional differences between batches of images. We show that on the Camelyon17 dataset, while both methods can adapt to global differences in color distribution, only our OT loss can reliably classify a cancer phenotype unseen during training. Together, our results suggest that OT improves generalization on rare but critical phenotypes that may only make up a small fraction of the total tiles and variation in a slide.

cs.CV↗

arXiv2021

Rightsizing the Railway Signal Workforce: a Zero-Based Resourcing Approach Towards Asset Management

Classic asset management approaches begin by inventorying all infrastructure assets and then assigning maintenance tasks and resources. Our approach collects similar data, but by starting with current personnel assignment and describing their job responsibilities and work processes, staff resistance in a railroad infrastructure owner-operator environment is minimized. Resulting "manning model" quantitatively measures signal maintenance burden including Federally mandated tests, trouble tickets, non-FRA maintenance, overhead and vacation coverage, location/shift assignment, administrative process, and work curfew productivity losses. It is capable of delivering immediate results by rightsizing allocation of workforce across shifts and maintenance base locations--even before all assets are formally inventoried. Typical data from a commuter passenger railroad shows that work curfews and shift assignment constraints have significant impacts on workforce productivity. Just over half of signal maintenance employee-hours are devoted to Federally mandated tests, whilst non-FRA and repair maintenance consumes about 25% each. These indicators provide intelligence driving strategic management actions to improve signal maintenance cost-effectiveness. This model provides workload-based employee assignment by craft, location, gang, and shift for maintenance manager use, but also provides analytical basis for establishing or abolishing positions in the budgeting process. Comparing its results with current employee payroll provides a measure of how much staffing stress the maintenance organization is under, which can help measure whether the current overtime usage is appropriate. Asset and maintenance task inventories collected in this process can also feed normal asset management processes to assess replacement cycles, asset failure risk, and to inform strategic and investment decisions.

eess.SY↗

arXiv2021

Power Off! Challenges in Planning and Executing Power Isolations on Shared-Use Electrified Railways

Electric railways are fast, clean, and safe, but complex to operate and maintain. Electric traction infrastructure includes signal power and feeder lines that remain live during isolations and complicate maintenance processes. Stakeholders involved in power outage planning include contractors, linemen, groundmen, power directors, dispatchers, conductor-flag, and support personnel. Weekly planning processes for track time requires many contingencies due to large number of moving parts and factors not known in advance, like personnel availability. Electrical and mechanical environments faced by crews working in adjacent areas may be entirely different and require a "bespoke" circuit configuration to de-energize catenary, which must be planned meticulously. Although recent automation improved real-time "plate order" communications between power directors and dispatchers, each outage still requires many manual switching operations. Net impact of this isolation process reduces available construction work windows nightly from a nominal 7 hours to 2 hrs 39 mins. We recommend joint design of electrical and civil infrastructure, cross-training between disciplines, limiting maximum number of concurrent outages, formal study of maintenance outage capacity, and further automation in power switching.

eess.SY↗

arXiv2020

Improved Conditional Flow Models for Molecule to Image Synthesis

In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development. Building on the recent success of graph neural networks for learning molecular embeddings and flow-based models for image generation, we propose Mol2Image: a flow-based generative model for molecule to cell image synthesis. To generate cell features at different resolutions and scale to high-resolution images, we develop a novel multi-scale flow architecture based on a Haar wavelet image pyramid. To maximize the mutual information between the generated images and the molecular interventions, we devise a training strategy based on contrastive learning. To evaluate our model, we propose a new set of metrics for biological image generation that are robust, interpretable, and relevant to practitioners. We show quantitatively that our method learns a meaningful embedding of the molecular intervention, which is translated into an image representation reflecting the biological effects of the intervention.

q-bio.BM↗