Search

We propose an AutoML system that enables model selection on clustering problems by leveraging optimal transport-based dataset similarity. Our objective is to establish a comprehensive AutoML pipeline for clustering problems and provide recommendations for selecting the most suitable algorithms, thus opening up a new area of AutoML beyond the traditional supervised learning settings. We compare our results against multiple clustering baselines and find that it outperforms all of them, hence demonstrating the utility of similarity-based automated model selection for solving clustering applications.

In distributed machine learning, efficient training across multiple agents with different data distributions poses significant challenges. Even with a centralized coordinator, current algorithms that achieve optimal communication complexity typically require either large minibatches or compromise on gradient complexity. In this work, we tackle both centralized and decentralized settings across strongly convex, convex, and nonconvex objectives. We first demonstrate that a basic primal-dual method, (Accelerated) Gradient Ascent Multiple Stochastic Gradient Descent (GA-MSGD), applied to the Lagrangian of distributed optimization inherently incorporates local updates, because the inner loops of running Stochastic Gradient Descent on the primal variable require no inter-agent communication. Notably, for strongly convex objectives, (Accelerated) GA-MSGD achieves linear convergence in communication rounds despite the Lagrangian being only linear in the dual variables. This is due to a structural property where the dual variable is confined to the span of the coupling matrix, rendering the dual problem strongly concave. When integrated with the Catalyst framework, our approach achieves nearly optimal communication complexity across various settings without the need for minibatches.

Class incremental learning (CIL) in an online continual learning setting strives to acquire knowledge on a series of novel classes from a data stream, using each data point only once for training. This is more realistic compared to offline modes, where it is assumed that all data from novel class(es) is readily available. Current online CIL approaches store a subset of the previous data which creates heavy overhead costs in terms of both memory and computation, as well as privacy issues. In this paper, we propose a new online CIL approach called FOCIL. It fine-tunes the main architecture continually by training a randomly pruned sparse subnetwork for each task. Then, it freezes the trained connections to prevent forgetting. FOCIL also determines the sparsity level and learning rate per task adaptively and ensures (almost) zero forgetting across all tasks without storing any replay data. Experimental results on 10-Task CIFAR100, 20-Task CIFAR100, and 100-Task TinyImagenet, demonstrate that our method outperforms the SOTA by a large margin. The code is publicly available at https://github.com/muratonuryildirim/FOCIL.

Effective operations and maintenance (O&M) in modern production systems hinges on careful orchestration of economic and degradation dependencies across a multitude of assets. While the economic dependencies are well studied, degradation dependencies and their impact on system operations remain an open challenge. To address this challenge, we model condition-based production and maintenance decisions for multi-asset systems with degradation interactions. There is a rich literature on condition-based O&M policies for single-asset systems. These models fail to represent modern systems composed of multiple interacting assets. We are providing the first O&M model to optimize O&M in multi-asset systems with embedded decision-dependent degradation interactions. We formulate robust optimization models that inherently capture degradation and failure risks by embedding degradation signals via a set of constraints, and building condition-based uncertainty sets to model probable degradation scenarios. We offer multiple reformulations and a solution algorithm to ensure computational scalability. Performance of the proposed O&M model is evaluated through extensive experiments, where the degradation is either emulated or taken from vibration-based readings from a rotating machinery system. The proposed model provides significant improvements in terms of operation, maintenance, and reliability metrics. Due to a myriad of dependencies across assets and decisions, it is often difficult to translate asset-level failure predictions to system-level O&M decisions. This challenge puts a significant barrier to the return on investment in condition monitoring and smart maintenance systems. Our approach offers a seamless integration of data-driven failure modeling and mathematical programming to bridge the gap across predictive and prescriptive models.

Battery diagnosis, prognosis and health management models play a critical role in the integration of battery systems in energy and mobility fields. However, large-scale deployment of these models is hindered by a myriad of challenges centered around data ownership, privacy, communication, and processing. State-of-the-art battery diagnosis and prognosis methods require centralized collection of data, which further aggravates these challenges. Here we propose a federated battery prognosis model, which distributes the processing of battery standard current-voltage-time-usage data in a privacy-preserving manner. Instead of exchanging raw standard current-voltage-time-usage data, our model communicates only the model parameters, thus reducing communication load and preserving data confidentiality. The proposed model offers a paradigm shift in battery health management through privacy-preserving distributed methods for battery data processing and remaining lifetime prediction.

Continual learning (CL) refers to the ability of an intelligent system to sequentially acquire and retain knowledge from a stream of data with as little computational overhead as possible. To this end; regularization, replay, architecture, and parameter isolation approaches were introduced to the literature. Parameter isolation using a sparse network which enables to allocate distinct parts of the neural network to different tasks and also allows to share of parameters between tasks if they are similar. Dynamic Sparse Training (DST) is a prominent way to find these sparse networks and isolate them for each task. This paper is the first empirical study investigating the effect of different DST components under the CL paradigm to fill a critical research gap and shed light on the optimal configuration of DST for CL if it exists. Therefore, we perform a comprehensive study in which we investigate various DST components to find the best topology per task on well-known CIFAR100 and miniImageNet benchmarks in a task-incremental CL setup since our primary focus is to evaluate the performance of various DST criteria, rather than the process of mask selection. We found that, at a low sparsity level, Erdos-R\'enyi Kernel (ERK) initialization utilizes the backbone more efficiently and allows to effectively learn increments of tasks. At a high sparsity level, unless it is extreme, uniform initialization demonstrates a more reliable and robust performance. In terms of growth strategy; performance is dependent on the defined initialization strategy and the extent of sparsity. Finally, adaptivity within DST components is a promising way for better continual learners.

Class-Incremental Learning aims to update a deep classifier to learn new categories while maintaining or improving its accuracy on previously observed classes. Common methods to prevent forgetting previously learned classes include regularizing the neural network updates and storing exemplars in memory, which come with hyperparameters such as the learning rate, regularization strength, or the number of exemplars. However, these hyperparameters are usually only tuned at the start and then kept fixed throughout the learning sessions, ignoring the fact that newly encountered tasks may have varying levels of novelty or difficulty. This study investigates the necessity of hyperparameter `adaptivity' in Class-Incremental Learning: the ability to dynamically adjust hyperparameters such as the learning rate, regularization strength, and memory size according to the properties of the new task at hand. We propose AdaCL, a Bayesian Optimization-based approach to automatically and efficiently determine the optimal values for those parameters with each learning task. We show that adapting hyperpararmeters on each new task leads to improvement in accuracy, forgetting and memory. Code is available at https://github.com/ElifCerenGokYildirim/AdaCL., Comment: Published in 1st ContinualAI Unconference

The rapid increase in scale and sophistication of offshore wind (OSW) farms poses a critical challenge related to the cost-effective operation and management of wind energy assets. A defining characteristic of this challenge is the economic trade-off between two concomitant processes: power production (the primary driver of short-term revenues), and asset degradation (the main determinant of long-term expenses). Traditionally, approaches to optimize production and maintenance in wind farms have been conducted in isolation. In this paper, we conjecture that a joint optimization of those two processes, achieved by rigorously modeling their short- and long-term dependencies, can unlock significant economic benefits for wind farm operators. In specific, we propose a decision-theoretic framework, rooted in stochastic optimization, which seeks a sensible balance of how wind loads are leveraged to harness short-term electricity generation revenues, versus alleviated to hedge against longer-term maintenance expenses. Extensive numerical experiments using real-world data confirm the superior performance of our approach, in terms of several operational performance metrics, relative to methods that tackle the two problems in isolation.

Limited throughput is a key challenge in in-vivo deep-tissue imaging using nonlinear optical microscopy. Point scanning multiphoton microscopy, the current gold standard, is slow especially compared to the wide-field imaging modalities used for optically cleared or thin specimens. We recently introduced 'De-scattering with Excitation Patterning or DEEP', as a widefield alternative to point-scanning geometries. Using patterned multiphoton excitation, DEEP encodes spatial information inside tissue before scattering. However, to de-scatter at typical depths, hundreds of such patterned excitations are needed. In this work, we present DEEP$^2$, a deep learning based model, that can de-scatter images from just tens of patterned excitations instead of hundreds. Consequently, we improve DEEP's throughput by almost an order of magnitude. We demonstrate our method in multiple numerical and physical experiments including in-vivo cortical vasculature imaging up to four scattering lengths deep, in alive mice.

Recent years have seen an unprecedented growth in the use of sensor data to guide wind farm operations and maintenance. Emerging sensor-driven approaches typically focus on optimal maintenance procedures for single turbine systems, or model multiple turbines in wind farms as single component entities. In reality, turbines are composed of multiple components that dynamically interact throughout their lifetime. These interactions are central for realistic assessment and control of turbine failure risks. In this paper, an integrated framework that combines i) real-time degradation models used for predicting remaining life distribution of each component, with ii) mixed integer optimization models and solution algorithms used for identifying optimal wind farm maintenance and operations is proposed. Maintenance decisions identify optimal times to repair every component, which in turn, determine the failure risk of the turbines. More specifically, optimization models that characterize a turbine's failure time as the first time that one of its constituent components fail - a systems reliability concept called competing risk is developed. The resulting turbine failures impact the optimization of wind farm operations and revenue. Extensive experiments conducted for multiple wind farms with 300 wind turbines - 1200 components - showcases the performance of the proposed framework over conventional methods., Comment: The paper has been accepted to Renewable and Sustainable Energy Reviews

Industrial sensor data provides significant insights into the failure risks of microgrid generation assets. In traditional applications, these sensor-driven risks are used to generate alerts that initiate maintenance actions without considering their impact on operational aspects. The focus of this paper is to propose a framework that i) builds a seamless integration between sensor data and operational & maintenance drivers, and ii) demonstrates the value of this integration for improving multiple aspects of microgrid operations. The proposed framework offers an integrated stochastic optimization model that jointly optimizes operations and maintenance in a multi-microgrid setting. Maintenance decisions identify optimal crew routing, opportunistic maintenance, and repair schedules as a function of dynamically evolving sensor-driven predictions on asset life. Operational decisions identify commitment and generation from a fleet of distributed energy resources, storage, load management, as well as power transactions with the main grid and neighboring microgrids. Operational uncertainty from renewable generation, demand, and market prices are explicitly modeled through scenarios in the optimization model. We use the structure of the model to develop a decomposition-based solution algorithm to ensure computational scalability. The proposed model provides significant improvements in reliability and enhances a range of operational outcomes, including costs, renewables, generation availability, and resilience., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

Decentralized methods are gaining popularity for data-driven models in power systems as they offer significant computational scalability while guaranteeing full data ownership by utility stakeholders. However, decentralized methods still require sharing information about network flow estimates over public facing communication channels, which raises privacy concerns. In this paper we propose a differential privacy driven approach geared towards decentralized formulations of mixed integer operations and maintenance optimization problems that protects network flow estimates. We prove strong privacy guarantees by leveraging the linear relationship between the phase angles and the flow. To address the challenges associated with the mixed integer and dynamic nature of the problem, we introduce an exponential moving average based consensus mechanism to enhance convergence, coupled with a control chart based convergence criteria to improve stability. Our experimental results obtained on the IEEE 118 bus case demonstrate that our privacy preserving approach yields solution qualities on par with benchmark methods without differential privacy. To demonstrate the computational robustness of our method, we conduct experiments using a wide range of noise levels and operational scenarios.

Unit Commitment (UC) is a fundamental problem in power system operations. When coupled with generation maintenance, the joint optimization problem poses significant computational challenges due to coupling constraints linking maintenance and UC decisions. Obviously, these challenges grow with the size of the network. With the introduction of sensors for monitoring generator health and condition-based maintenance(CBM), these challenges have been magnified. ADMM-based decentralized methods have shown promise in solving large-scale UC problems, especially in vertically integrated power systems. However, in their current form, these methods fail to deliver similar computational performance and scalability when considering the joint UC and CBM problem. This paper provides a novel decentralized optimization framework for solving large-scale, joint UC and CBM problems. Our approach relies on the novel use of the subgradient method to temporally decouple various subproblems of the ADMM-based formulation of the joint problem along the maintenance horizon. By effectively utilizing multithreading, our decentralized subgradient approach delivers superior computational performance and eliminates the need to move sensor data thereby alleviating privacy and security concerns. Using experiments on large scale test cases, we show that our framework can provide a speedup of upto 50x as compared to various state of the art benchmarks without compromising on solution quality.

We report on the nonlinear spectroscopy of water vapor at THz frequencies. Atmospheric water vapor has a rich spectrum with several strong resonances at frequencies below 3 THz, falling within the range of operation of most existing THz sources. We observe an extremely large nonlinear response to THz radiation at the positions of these resonances. Using the optical Kerr model for the nonlinear response, we estimate a minimum nonlinear refractive index of the order of 10^2 m^2/W. Our results provide insight into the energy levels of the water molecule and give a more accurate picture of its response to electromagnetic radiation, paving the way to more accurate THz spectroscopy, imaging and sensing systems, and thereby facilitating future emerging THz technologies.

Initial development and subsequent calibration of discrete event simulation models for complex systems require accurate identification of dynamically changing process characteristics. Existing data driven change point methods (DD-CPD) assume changes are extraneous to the system, thus cannot utilize available process knowledge. This work proposes a unified framework for process-driven multi-variate change point detection (PD-CPD) by combining change point detection models with machine learning and process-driven simulation modeling. The PD-CPD, after initializing with DD-CPD's change point(s), uses simulation models to generate system level outputs as time-series data streams which are then used to train neural network models to predict system characteristics and change points. The accuracy of the predictive models measures the likelihood that the actual process data conforms to the simulated change points in system characteristics. PD-CPD iteratively optimizes change points by repeating simulation and predictive model building steps until the set of change point(s) with the maximum likelihood is identified. Using an emergency department case study, we show that PD-CPD significantly improves change point detection accuracy over DD-CPD estimates and is able to detect actual change points., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

With increased reliance on cyber infrastructure, large scale power networks face new challenges owing to computational scalability. In this paper we focus on developing an asynchronous decentralized solution framework for the Unit Commitment(UC) problem for large scale power networks. We exploit the inherent asynchrony in a region based decomposition arising out of imbalance in regional subproblems to boost computational efficiency. A two phase algorithm is proposed that relies on the convex relaxation and privacy preserving valid inequalities in order to deliver algorithmic improvements. Our algorithm employs a novel interleaved binary mechanism that locally switches from the convex subproblem to its binary counterpart based on consistent local convergent behavior. We develop a high performance computing (HPC) oriented software framework that uses Message Passing Interface (MPI) to drive our benchmark studies. Our simulations performed on the IEEE 3012 bus case are benchmarked against the centralized and a state of the art synchronous decentralized method. The results demonstrate that the asynchronous method improves computational efficiency by a significant amount and provides a competitive solution quality rivaling the benchmark methods.