A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Paper optimizes FL communication efficiency with stochastic optimization.
NeuralPrefix fills in missing sensor data without additional training.
Paper shows local SGD outperforms mini-batch SGD under certain conditions.
Paper addresses state estimation in sensor networks with intermittent data.
Federated Learning is a powerful machine learning paradigm to cooperatively train a global model with highly distributed data. A major bottleneck on the performance of distributed Stochastic Gradient Descent (SGD) algorithm for large-scale Federated Learning is the communication overhead on pushing local gradients and …
New algorithm reduces sample and communication complexities in federated Q-learning.
This article proposes a communication-efficient decentralized deep learning algorithm, coined layer-wise federated group ADMM (L-FGADMM). To minimize an empirical risk, every worker in L-FGADMM periodically communicates with two neighbors, in which the periods are separately adjusted for different layers of its deep ne…
This paper introduces intermittent learning - the goal of which is to enable energy harvested computing platforms capable of executing certain classes of machine learning tasks effectively and efficiently. We identify unique challenges to intermittent learning relating to the data and application semantics of machine l…
We suggest a general oracle-based framework that captures different parallel stochastic optimization settings described by a dependency graph, and derive generic lower bounds in terms of this graph. We then use the framework and derive lower bounds for several specific parallel optimization settings, including delayed …
The study uses Gaussian Processes with Tweedie likelihood for forecasting intermittent time series.
Intermittent demand, where demand occurrences appear sporadically in time, is a common and challenging problem in forecasting. In this paper, we first make the connections between renewal processes, and a collection of current models used for intermittent demand forecasting. We then develop a set of models that benefit…
This paper analyzes Local SGD for federated learning, achieving both statistical and communication efficiency.
Study on collaborative vs. non-collaborative online and bandit convex optimization.
Global models outperform local models in forecasting intermittent time series.
Unified framework for intermittent demand forecasting using renewal processes.
In this empirical paper, we investigate how learning agents can be arranged in more efficient communication topologies for improved learning. This is an important problem because a common technique to improve speed and robustness of learning in deep reinforcement learning and many other machine learning algorithms is t…
A new Bayesian model improves forecasting for intermittent demand.
Federated learning algorithm improves with intermittent client availability.
Fault-tolerant federated learning for non-uniform data.
Maximum likelihood estimation applied to high-frequency data allows us to quantify intermittency in the fluctu- ations of asset prices. From time records as short as one month these methods permit extraction of a meaningful intermittency parameter λ characterising the degree of volatility clustering of asset prices. We…
Large-scale distributed training requires significant communication bandwidth for gradient exchange that limits the scalability of multi-node training, and requires expensive high-bandwidth network infrastructure. The situation gets even worse with distributed training on mobile devices (federated learning), which suff…
We present a model of financial markets originally proposed for a turbulent flow, as a dynamic basis of its intermittent behavior. Time evolution of the price change is assumed to be described by Brownian motion in a power-law potential, where the `temperature' fluctuates slowly. The model generally yields a fat-tailed…
Kaggle competitions offer valuable insights for business forecasting.
A new metric optimizes forecasts for lumpy, intermittent demand.
Minibatch SGD outperforms Local SGD in heterogeneous distributed learning.
A new model captures multifractal volatility in stock returns.
A new model captures multifractal volatility in stock returns.
Factorial moments are convenient tools in nuclear physics to characterize the multiplicity distributions when phase-space resolution () becomes small. For uncorrelated particle production within , Gaussian statistics holds and factorial moments are equal to unity for all orders . Correlations between par…
In this paper we study the price dynamics in a simple model of financial markets with heterogeneous agents. We concentrate on how increases in the total number of active traders influences fluctuations of asset prices. We find that a curious route to chaos is observed when the total number of [active traders] increases…
New algorithm reduces communication time in federated learning.
Study uses AUVs and RL to map river plumes over multiple days.
New method for decentralized learning reduces data and computation needs.
Multiplicative random cascade model naturally reproduces the intermittency or multifractality, which is frequently shown among hierarchical complex systems such as turbulence and financial markets. As described herein, we investigate the validity of a multiplicative hierarchical random cascade model through an empirica…
We present a scalable and robust Bayesian inference method for linear state space models. The method is applied to demand forecasting in the context of a large e-commerce platform, paying special attention to intermittent and bursty target statistics. Inference is approximated by the Newton-Raphson algorithm, reduced t…
Sharp heat kernel estimates on manifolds lead to solutions of the Parabolic Anderson model.
SurvSurf predicts first hitting times for intermittent events without monotonic violations.
A new asynchronous method for vertical federated learning improves privacy and efficiency.
Proposes a pricing agent using reinforcement learning to balance renewable energy demand.
A new hierarchical forecasting method improves overall accuracy.
The Minority Game framework was recently generalized to account for the possibility that agents adapt not only through strategy selection but also by diversifying their response according to the kind of dynamical regime, or the risk, they perceive. Here we study the effects of this mechanism in different information st…
Successful implementation of California's Renewable Portfolio Standard (RPS) mandating 33 percent renewable energy generation by 2020 requires inclusion of a robust strategy to mitigate increased risk of energy deficits (blackouts) due to short time-scale (sub 1 hour) intermittencies in renewable energy sources. Of the…
Studying the impact of climate change on precipitation is constrained by finding a way to evaluate the evolution of precipitation variability over time. Classical approaches (feature-based) have shown their limitations for this issue due to the intermittent and irregular nature of precipitation. In this study, we prese…
Unified analysis of FL with arbitrary client participation.
We tackle anomaly detection in sparse time series data.
ACI uses Bayesian data assimilation to trace causes from effects in complex systems.
In this paper, we show how simple logistic growth that was studied intensively during the last 200 years in many domains of science could be extended in a rather simple way and with these extensions is capable to produce a collection of behaviors widely observed in an enormous number of real-life systems in Economics, …
The formation of price in a financial market is modelled as a chain of Ising spin with three fundamental figures of trading. We investigate the time behaviour of the model, and we compare the results with the real EURO/USD change rate. By using the test of local Poisson hypothesis, we show that this minimal model leads…
New method models complex dynamics using a base variable.