New Async-SGD and Async-SGDI methods converge for non-convex problems with unbounded delays.
problem Improving convergence of asynchronous stochastic gradient descent with unbounded delays in non-convex learning.
method Developed Async-SGD and Async-SGDI methods for non-convex optimization with unbounded gradient delays, proving convergence rates and establishing a unifying sufficient condition.
result Proved o ( 1 / k ) o(1/\sqrt{k}) o ( 1/ k ) convergence rate for Async-SGD and o ( 1 / k ) o(1/k) o ( 1/ k ) for Async-SGDI. This paper analyzes async-parallel algorithms with unbounded delays, proving convergence and providing a stepsize formula.
problem Problems with asynchrony in parallel iterations and unbounded delays.
method Probabilistic analysis of async-parallel methods with large unbounded delays, providing an explicit stepsize formula.
result An explicit formula for stepsize that guarantees convergence under large unbounded delays.
New algorithm tackles delayed feedback in Lipschitz bandits with sublinear regret.
problem Delayed feedback in Lipschitz bandits.
method Design of algorithms for bounded and unbounded stochastic delays.
result Sublinear regret guarantees for both bounded and unbounded delays.
New metrics needed for streaming ML due to delayed labels.
problem Streaming ML evaluation fails to identify unexpected performance.
method Recommend additional metrics for streaming ML performance.
result New metrics are needed for streaming ML due to delayed labels.
New bandit problem with delayed, aggregated feedback analyzed.
problem Stochastic K K K -armed bandit problem with delayed, aggregated anonymous feedback. method Developed algorithm matching worst case regret of non-anonymous problem.
result Regret increase can be maintained in the harder delayed, aggregated anonymous feedback setting.
DASA speeds up SA with delayed agents, achieving N-fold speedup.
problem Speeding up Stochastic Approximation with asynchronous delays.
method DASA: Delay-Adaptive Multi-Agent Stochastic Approximation algorithm.
result First algorithm with convergence rate dependent on mixing time and average delay.
New algorithm for multiarmed bandits with variable, unbounded delays achieves similar regret bounds.
problem Variable, unbounded delays in multiarmed bandits.
method Introduces a new algorithm that skips rounds with excessively large delays and uses a doubling scheme.
result Achieves the same regret bound as Exp3 with variable, unbounded delays.
Paper develops accelerated APCD for nonconvex nonsmooth problems with performance guarantees.
problem Efficient methods for nonconvex nonsmooth optimization problems with performance guarantees.
method Asynchronous Accelerated Proximal Coordinate Descent (AAPCD) for nonsmooth and nonconvex problems.
result AAPCD ensures that every limit point is a critical point and achieves linear and sublinear convergence rates.
PITMonitor monitors model calibration over time with formal error guarantees.
problem Fixed-sample tests applied to models over time can lead to false alarms.
method PITMonitor uses mixture e-processes to detect distributional shifts in probability integral transforms.
result PITMonitor achieves competitive detection rates on river's FriedmanDrift benchmark.
FedBuff improves federated learning scalability with asynchronous updates.
problem Limited scalability of federated learning with synchronous updates.
method Introduces asynchronous updates (staleness) in federated learning.
result Theoretical analysis shows improved convergence rate with boundedness removed.
Develops a stochastic approach to financial market delays.
problem Modeling delays in financial markets with multiple assets.
method Introduces a general stochastic framework for information and order execution delays.
result Delayed markets maintain fundamental asset pricing theorems and no asymptotic free lunch condition.
Paper tackles action delays in reinforcement learning, proposing a delay-aware framework.
problem Action delays degrade reinforcement learning performance in real-world systems.
method Formal definition of delay-aware MDP, transformation into standard MDP with augmented states, delay-aware model-based reinforcement learning framework.
result Proposed framework is more efficient in training and transferable between systems with various delay durations.
New algorithms ensure fair selection in combinatorial semi-bandit with unrestricted delays.
problem Fair selection in stochastic combinatorial semi-bandit with delayed feedback.
method Introduced merit-based fairness constraints and new bandit algorithms for reward and fairness.
result Achieved sublinear expected reward and fairness regrets with dependence on delay distribution quantiles.
Banker-OMD improves online learning with delayed feedback.
problem Handling delayed feedback in online learning.
method Generalized Online Mirror Descent (OMD) framework.
result Achieves nearly-optimal performance in three bandit scenarios.
New algorithm handles delayed feedback robustly, reducing regret without knowing delay bounds.
problem Bandits with variably delayed feedback, especially excessive delays.
method Implicit exploration scheme, adaptive skipping, drifted regret control.
result Can tolerate arbitrary excessive delays up to order T, reducing regret.
Proposes a nonparametric model for predicting conversion rates with delayed feedback.
problem Predicting conversion rates with time delays and unknown distribution.
method Nonparametric delayed feedback model without assuming a specific distribution.
result The proposed model outperforms existing methods in conversion rate prediction.
Gradient descent with delayed updates converges faster with noise, even when delays are significant.
problem Analyzing convergence of gradient descent with delayed gradients and stochastic noise.
method Novel technique using generating functions for convergence analysis.
result Convergence bounds show that stochastic noise mitigates the negative effects of delays, improving performance.
Paper tackles delays in multi-agent reinforcement learning, improving performance.
problem Challenges in reinforcement learning due to delays in real-world systems.
method Proposes a novel framework for multi-agent reinforcement learning with delays, using Delay-Aware Markov Games and centralized-decentralized training.
result Demonstrates significant improvement in performance with delay-aware multi-agent reinforcement learning.
Derives a Feynman-Kac formula for a fixed delay CIR model.
problem Modeling financial processes with fixed delay.
method Proves existence and uniqueness of a strong solution for a specific SDDE.
result Derives a Feynman-Kac type formula leading to an affine bond pricing formula.
Study on synchronization in financial markets with time delays.
problem Understanding market dynamics and synchronization in financial systems with time delays.
method Examined a system of coupled non-linear delay-differential equations, linearized for small delays, and analyzed collective dynamics using bifurcation diagrams and numerical solutions.
result Demonstrated that limit cycles can be maintained in coupled N-asset models with appropriate parameterization, leading to market synchronization.
BayTiDe discovers time-delayed differential equations from noisy data.
problem Discovering time-delayed differential equations from data with large delays and noise.
method Bayesian inference with a sparsity-promoting prior.
result BayTiDe accurately identifies time-delayed differential equations with accuracy proportional to data resolution.
New algorithm tackles stochastic bandits with varying arm-dependent delays.
problem Applying existing algorithms to stochastic delayed bandit settings is restricted by strong assumptions on delay distributions.
method Proposes a simple UCB-based algorithm called PatientBandits that weakens assumptions on delay distributions.
result Provides bounds on regret and performance lower bounds for the PatientBandits algorithm.
Model analyzes how delayed information impacts option pricing.
problem Effects of delayed information on option pricing.
method Binomial model, closed form formula for convex contingent claims, convergence analysis.
result Delayed information exaggerates the volatility smile.
TSMB handles time delays in multivariate time series data.
problem Varying time delays in multivariate time series data complicate predictions.
method Time Series Model Bootstrap (TSMB) framework for nonparametric time delay estimation.
result TSMB improves model performance in dynamic data environments.
Delayed-RNN approximates stacked and bidirectional RNNs.
problem Improving RNN expressiveness and representational capacity.
method Weight-constrained delayed-RNN, equivalent to stacked-RNNs, with partial acausality.
result Delayed-RNN can approximate stacked and bidirectional RNNs, outperforming them in some tasks.
New algorithm reduces regret in delayed feedback generalised linear bandits.
problem Regret in delayed feedback generalised linear bandits.
method Adaptation of optimistic algorithm to delayed feedback.
result Achieves a regret bound independent of the horizon's delay penalty.
Adapts Exp3 to adversarial bandits with delays and data.
problem Adversarial multi-armed bandits with delayed feedback.
method Tuned Exp3 variants with step-size adaptation and implicit exploration.
result Optimal regret bounds of log ( K ) ( T K + D ) \sqrt{\log(K)(TK + D)} log ( K ) ( T K + D ) with high probability. Capacity-Constrained Online Convex Optimization with Delayed Feedback
problem Online learning with delayed feedback under a hard capacity constraint
method Reduction to a delayed and weighted OCO problem using a scheduler
result First regret guarantees for capacity-constrained OCO under convex and strongly convex losses
New algorithm tackles non-stationary delayed feedback in recommender systems.
problem Challenges in learning from delayed feedback in non-stationary environments.
method Developed a UCRL-based algorithm for non-stationary, delayed bandits with intermediate observations.
result Sublinear regret guarantees for the proposed algorithm in non-stationary delayed environments.
PCTS optimizes noisy, delayed, multi-fidelity feedbacks in black-box optimization.
problem Optimizing unknown functions with noisy, delayed, and multi-fidelity feedbacks.
method ProCrastinated Tree Search (PCTS) with DUCB1 and DUCBV algorithms.
result PCTS achieves better regret bounds for delayed, noisy, and multi-fidelity feedbacks.
Study online learning with delays and capacity constraints, achieving optimal regret bounds.
problem Online learning with delays and capacity constraints.
method Novel scheduling and preemptive techniques, matching upper and lower bounds.
result Achieves optimal regret bounds across all capacity levels.
Study market delay effects on contingent claims pricing.
problem Delayed market information impacts contingent claims pricing.
method Analyzes Black-Scholes and binomial models with delay.
result Scaling limit of super-replication prices equals G-expectation.
Online learning with delayed feedback has received increasing attention recently due to its several applications in distributed, web-based learning problems. In this paper we provide a systematic study of the topic, and analyze the effect of delay on the regret of online learning algorithms. Somewhat surprisingly, it t…
Federated learning technique improves convergence speed with communication delays.
problem Communication delays between edge nodes and aggregator in federated learning.
method Developed FedDelAvg, a technique that generalizes federated averaging to incorporate a weighting between current local model and delayed global model.
result FedDelAvg achieves a significant improvement in convergence speed, especially when optimizing the weighting scheme to account for delays.
Study uses randomized allocation for delayed rewards in multi-armed bandits.
problem Delayed rewards in contextual multi-armed bandits.
method Randomized allocation with nonparametric estimation.
result Strongly consistent strategy for delayed rewards.
New algorithms handle unknown delays in online learning problems.
problem Handling unknown delays in bandit online learning.
method DEXP3 for MAB and DBGD for BCO, with a unified analysis.
result Regret bounds of O ( K d ˉ ( T + D ) ) {\cal O}\big( \sqrt{K\bar{d}(T+D)} \big) O ( K d ˉ ( T + D ) ) for DEXP3 and O ( K ( T + D ) ) {\cal O}\big( \sqrt{K(T+D)} \big) O ( K ( T + D ) ) for DBGD. Develops hedging algorithm for online expert weight allocation with delayed feedback.
problem Adaptive hedging strategies for online expert weight allocation with delayed feedback.
method General Hedging algorithm G \mathcal{G} G based on exponential reweighing of experts' losses. result Proves adversarial loss bounds for the General Hedging algorithm G \mathcal{G} G in the delayed feedback setting. Study a dual risk model with innovation delays, focusing on ruin probability and time.
problem Analyzing risk models with innovation delays affecting ruin probability and time.
method Delayed dual risk model with innovation delays, focusing on ruin probability and time.
result Closed-form formulas for ruin probability and time in some special cases.
Simple regret bound for online optimization with adversarial delays.
problem Online strongly-convex optimization with adversarial delays.
method Online Gradient Descent algorithm with a specific regret bound.
result Simple regret bound of \Oh{\sum_{t=1}^T \log (1+ \frac{d_t}{t})}
Study on unimodality of plucking polynomial with delay function.
problem Exploring unimodality of plucking polynomial with delay function.
method Presented a formula for the plucking polynomial of hedgehog rooted trees and explored unimodality with specific delay functions.
result Found interesting examples and speculations on unimodality of plucking polynomials with delay functions.
Delay Pruning regularizes DyBM for better learning of temporal sequences.
problem Improving DyBM's generalization in learning high-dimensional temporal sequences.
method Delay Pruning: setting some FIFO queue lengths to one with fixed probability.
result Delay Pruning enhances DyBM's performance in learning high-dimensional temporal sequences.
New algorithms reduce forecasting errors by leveraging optimistic learning and hinting.
problem Real-time climate forecasting with delayed feedback.
method Developed optimistic online learning algorithms that reduce regret under delayed feedback.
result Optimistic learning algorithms achieve optimal regret guarantees under delayed feedback.
A new algorithm tackles delayed combinatorial semi-bandit with causal relations.
problem Optimizing decisions in a non-stationary environment with delayed and causally related rewards.
method Formalized as a non-stationary delayed combinatorial semi-bandit problem, the approach models causal relations with a directed graph in a stationary structural equation model. The agent learns these relations from delayed feedback to optimize decisions.
result Proved a regret bound for the proposed algorithm's performance.
The paper models delayed event occurrences in insurance data.
problem Systematic underestimation of event occurrence due to observation delays.
method Modeling the time between event occurrence and observation, considering event day and calendar effects.
result A granular model for event observation delay heterogeneity.
New method estimates traffic congestion delays using statistical causality.
problem Accurate estimation of traffic congestion delays during accidents.
method Proposes a novel time delay estimation method using lag-specific transfer entropy (TE) and Markov bootstrap techniques.
result Validated the method's efficacy using simulated and real data.
New method solves stochastic control problems with delays using deep learning.
problem Stochastic control problems with delayed control in drift and diffusion.
method Characterization via Riccati PDEs and deep learning scheme.
result Illustrates effect of delay on Markowitz portfolio allocation problem.
Adapts two algorithms for online learning with delayed rewards.
problem Online learning with delayed rewards in generalized linear contextual bandits.
method Modifies upper confidence bounds and Thompson sampling algorithms for delayed rewards.
result Both algorithms can be made robust to delays, improving their performance.
Neural Laplace Control tackles offline RL for continuous-time delayed systems with irregular observations.
problem Offline reinforcement learning problems involving continuous-time environments with delays and irregular observations.
method Combines a Neural Laplace dynamics model with a model predictive control (MPC) planner.
result Achieves near expert policy performance on continuous-time delayed environments.