Central Bank Digital Currencies and Banking Stability: Evidence on Monetary Transmission from Emerging Economies

Zenodo (CERN European Organization for Nuclear Research) · 2026-01-01

Purpose – This study examines how the introduction of Central Bank Digital Currencies (CBDCs)affects banking sector stability and the effectiveness of monetary policy transmission in emergingeconomies.Methodology – Using a panel dataset of countries that have piloted or adopted CBDCs, wecombine banking sector indicators, deposit flows, and monetary policy variables. A difference-indifferences framework is applied, supplemented with dynamic panel regressions to ensurerobustness.Findings – The results show that CBDCs strengthen monetary policy transmission by improvingthe pass-through of policy rates to lending markets. However, they also exert short-term pressureon bank deposit bases, with effects more pronounced in economies with weaker financialregulation. These dual outcomes suggest that CBDCs act simultaneously as stabilizers anddisruptors within the banking system.Implications – The findings highlight the need for regulatory safeguards and adaptive policyframeworks to mitigate risks of disintermediation while leveraging the efficiency gains of CBDCs.Emerging economies must design CBDCs carefully to balance innovation with financial stability.

Central Bank Digital Currencies and Banking Stability: Evidence on Monetary Transmission from Emerging Economies

Zenodo (CERN European Organization for Nuclear Research) · 2026-01-01

Purpose – This study examines how the introduction of Central Bank Digital Currencies (CBDCs)affects banking sector stability and the effectiveness of monetary policy transmission in emergingeconomies.Methodology – Using a panel dataset of countries that have piloted or adopted CBDCs, wecombine banking sector indicators, deposit flows, and monetary policy variables. A difference-indifferences framework is applied, supplemented with dynamic panel regressions to ensurerobustness.Findings – The results show that CBDCs strengthen monetary policy transmission by improvingthe pass-through of policy rates to lending markets. However, they also exert short-term pressureon bank deposit bases, with effects more pronounced in economies with weaker financialregulation. These dual outcomes suggest that CBDCs act simultaneously as stabilizers anddisruptors within the banking system.Implications – The findings highlight the need for regulatory safeguards and adaptive policyframeworks to mitigate risks of disintermediation while leveraging the efficiency gains of CBDCs.Emerging economies must design CBDCs carefully to balance innovation with financial stability.

Efficient SoC Power Estimation With Machine Learning

IEEE Transactions on Very Large Scale Integration (VLSI) Systems · 2026-02-24

We propose machine learning (ML) power, the first ML-based framework that: 1) accelerates end-to-end RTL power estimation by addressing both its key bottlenecks—RTL simulation and power-model evaluation—and 2) extends ML-based power models to system-on-chips (SoCs) with configurable intellectual property (IP) blocks. ML-Power builds models that predict power versus time traces of each SoC block using a very small subset of internal signals called power proxies. The framework is composed of three components: ML-based power proxy activity estimation (PACE), exploiting spatio-temporal correlations for power proxy selection and ML-based power model evaluation (SCOPE), and representative configuration selection using active learning (RECAL). PACE trains a sequence-to-sequence (seq2seq) ML model to translate a transaction-level execution trace into a cycle-level trace of the power proxy signals, allowing much faster simulation models to be used in place of RTL simulation. SCOPE selects power proxies and trains ML-based power models for each block within the SoC. RECAL enables ML-Power to handle SoCs with configurable intellectual property (IP) blocks by using active learning to select a small subset of representative configurations, which are then used to train a unified power model that generalizes across the entire design space. We evaluate ML-Power on an ARM SSE-300 SoC and two RISC-V-based SoCs. Compared to prior state-of-the-art ML-based power estimation frameworks, SCOPE trains power models <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 65\times $</tex-math> </inline-formula> faster, picks 40% fewer proxies, and achieves 2% lower estimation error. For the two RISC-V SoCs with configurable IP blocks, ML-Power achieves less than 10% error in per-cycle power across the entire design space using only 10–14 training configurations. ML-Power also achieves <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 300\times $</tex-math> </inline-formula> inference-time speedup over a commercial RTL power estimation tool with less than 7% error in per-cycle power estimates.

Challenges in Economic Policy Formation with Digital Market Disruptions: With Reference to India

Journal of Cultural Analysis and Social Change · 2026-02-22

India's digital economy is now among the most vibrant globally, adding a projected USD 500 billion to GDP in 2023, or 11% of output. The explosive expansion of fintech, e-commerce, and the gig economy holds opportunities but also challenges of regulation for policymakers. This paper combines quantitative analysis with qualitative examination to analyze the economic effects of digital market disruption. Based on secondary data from MeitY, RBI, World Bank, and NITI Aayog (2015–2024), the analysis reports strong positive correlations of digital adoption and GDP growth (r = 0.82), diversification of employment, and financial inclusion. Nevertheless, taxation, data privacy, and labour protection issues remain. The paper suggests evidence-based policies balancing innovation with equity to ensure long-term digital advancement.

ML-Enhanced Performance and Power Estimation for DNNs on Heterogenous SoCs

2025-09-08

Efficient deployment of Deep Neural Networks (DNNs) on heterogeneous Systems-on-Chip (SoCs) requires navigating a large and complex design space defined by combinations of choices in network architectures, operator-to-processor mappings, and power/performance operating points for each processor. Exploring this vast space requires estimators that are faster than even real-time execution on the target hardware. Machine learning (ML)-based estimators have been proposed to address this challenge. However, prior ML-based methods fail to accurately capture all the effects of heterogeneous parallel execution such as parallelism across processors, inter-processor communication, and control-flow dependencies from offloading, resulting in significant estimation errors.We introduce CoCO-ML, a Concurrency, Communication, and Offload-aware framework that integrates ML models with classical graph-based algorithms for fast and accurate performance and power estimation. CoCO-ML constructs a Computation-Communication Graph (CCG) that explicitly models operator mappings, data transfers between processors, and offload overheads. ML models are used to predict the execution time and power of individual operators, while symbolic graph traversal produces execution traces and end-to-end execution time and power estimates.We evaluate CoCO-ML on four single-DNN and two multi-DNN workloads executed on the NVIDIA Jetson AGX SoC. CoCO-ML achieves mean estimation errors of 6.54% (performance) and 12.51% (power), outperforming prior methods while being over 1000× faster than native execution. Furthermore, CoCO-ML generates symbolic execution traces that expose application bottlenecks, providing actionable insights to application developers and facilitating efficient deployment of DNNs on heterogeneous SoCs.

(LiFT) Lightweight Fitness Transformer: A language-vision model for Remote Monitoring of Physical Training

Smart Health · 2025-09-10 · 3 citations

We introduce a fitness tracking system that enables remote monitoring for exercises using only a RGB smartphone camera, making fitness tracking more private, scalable, and cost effective. Although prior work explored automated exercise supervision, existing models are either too limited in exercise variety or too complex for real-world deployment. Prior approaches typically focus on a small set of exercises and fail to generalize across diverse movements. In contrast, we develop a robust, multitask motion analysis model capable of performing exercise detection and repetition counting across hundreds of exercises, a scale far beyond previous methods. We overcome previous data limitations by assembling a large-scale fitness dataset, Olympia , covering more than 1,900 exercises. To our knowledge, our vision-language model is the first that can perform multiple tasks on skeletal fitness data. On Olympia , our model can detect exercises with 76.5% accuracy and count repetitions with 85.3% off-by-one accuracy, using only RGB video. By presenting a single vision-language transformer model for both exercise identification and rep counting, we take a significant step towards democratizing AI-powered fitness tracking.

Real-Time Heterogeneous Collaborative Perception in Edge-Enabled Vehicular Environments

IEEE Open Journal of Vehicular Technology · 2025-01-01 · 1 citations

Vehicular sensing has reached new heights due to advances in external perception systems enabled by the increasing number and type of sensors in vehicles, as well as the availability of on-board computing. These changes have led to improvements in driver safety and have also created a highly heterogeneous environment of vehicles on the road today in terms of sensing and computing. Using collaborative perception, the information obtained by vehicles with sensing capabilities can be expanded and improved, and older vehicles that lack external sensors and computing capabilities can be informed of potential hazards, opening the opportunity to improve traffic efficiency and safety on the roads. However, achieving real-time collaborative perception is a difficult task due to the dynamic availability of vehicular sensing and computing and the highly variable nature of vehicular communications. To address these challenges, we propose a Heterogeneous Adaptive Collaborative Perception (HAdCoP) framework which utilizes a Context-aware Latency Prediction Network (CaLPeN) to intelligently select which vehicles should transmit their sensor data, the specific individual and collaborative perception tasks, and the amount of computational offloading that should be utilized given information about the current state of the environment. Additionally, we propose an Adaptive Perception Frequency (APF) model to determine the optimal end-to-end latency requirement according to the current state of the environment. The proposed CaLPeN model outperforms six implemented comparison models in terms of effective mean average precision (EmAP), beating the next best model's performance by 5.5% on average when tested on the OPV2V perception dataset using two different combinations of wireless communication conditions and vehicular sensor/computing distributions.

(LiFT) Lightweight Fitness Transformer: A language-vision model for Remote Monitoring of Physical Training

ArXiv.org · 2025-06-06

We introduce a fitness tracking system that enables remote monitoring for exercises using only a RGB smartphone camera, making fitness tracking more private, scalable, and cost effective. Although prior work explored automated exercise supervision, existing models are either too limited in exercise variety or too complex for real-world deployment. Prior approaches typically focus on a small set of exercises and fail to generalize across diverse movements. In contrast, we develop a robust, multitask motion analysis model capable of performing exercise detection and repetition counting across hundreds of exercises, a scale far beyond previous methods. We overcome previous data limitations by assembling a large-scale fitness dataset, Olympia covering more than 1,900 exercises. To our knowledge, our vision-language model is the first that can perform multiple tasks on skeletal fitness data. On Olympia, our model can detect exercises with 76.5% accuracy and count repetitions with 85.3% off-by-one accuracy, using only RGB video. By presenting a single vision-language transformer model for both exercise identification and rep counting, we take a significant step toward democratizing AI-powered fitness tracking.

JExplore: Design Space Exploration Tool for Nvidia Jetson Boards

ArXiv.org · 2025-02-16

Nvidia Jetson boards are powerful systems for executing artificial intelligence workloads in edge and mobile environments due to their effective GPU hardware and widely supported software stack. In addition to these benefits, Nvidia Jetson boards provide large configurability by giving the user the choice to modify many hardware parameters. This large space of configurability creates the need of searching the optimal configurations based on the user's requirements. In this work, we propose JExplore, a multi-board software and hardware design space exploration tool. JExplore can be integrated with any search tool, hence creating a common benchmarking ground for the search algorithms. Moreover, it accelerates the exploration of user application and Nvidia Jetson configurations for researchers and engineers by encapsulating host-client communication, configuration management, and metric measurement.

ML-Power: Machine Learning based Power Estimation for SoCs

2025-08-06

Accurate power estimation early in the design cycle is crucial for the design of power-efficient System-on-Chips (SoCs). Power estimation has been researched at various levels of abstraction, with a well-known trade-off between efficiency and accuracy. Recent work has shown great promise for machine learning (ML) techniques to advance the state-of-the-art in power estimation. We propose ML-Power, the first ML-based framework that can address both key bottlenecks involved in power estimation, viz. simulation and power model evaluation. ML-Power builds and uses models that estimate power in each SoC block using a very small subset of internal signals, known as power proxies. ML-Power consists of two key components, PACE and SCOPE. PACE trains a sequence-to-sequence ML model to translate a transaction-level execution trace into a cycle-level trace of the power proxy signals, allowing much faster simulation models to be used in place of RTL simulation. SCOPE selects power proxies and trains ML-based power models for blocks (IPs) within the SoC. SCOPE improves upon prior work in ML-based power modeling by exploiting spatio-temporal correlations among SoC blocks to minimize the number of power proxies while also improving estimation accuracy. We evaluate ML-Power for an ARM SSE-300 SoC and two RISC-V based SoCs. Compared to prior state-of-the-art ML-based power estimation frameworks, SCOPE trains power models ∼65× faster, picks 40% fewer proxies, and achieves 2% lower estimation error. ML-Power also achieves ∼300× speedup over a commercial RTL power estimation tool with less than 7% error in per-cycle power estimates.