MVFST-RL tackles real-time network congestion control with RL.
problem Real-time network congestion control with RL in asynchronous environments.
method Asynchronous RL framework for delayed actions in QUIC.
result Scalable framework for congestion control in QUIC with improved state-of-the-art RL.
Unified framework for finite-sample RL algorithms using Lyapunov theory.
problem Finite-sample convergence guarantees of asynchronous RL algorithms.
method Reformulate RL algorithms as Markovian SA, develop Lyapunov analysis.
result Mean-square error bounds and convergence for various RL algorithms.
Catalyst.RL accelerates RL research with efficient training.
problem Efficient reinforcement learning training in complex environments.
method Open-source PyTorch framework with distributed training and RL algorithms.
result Catalyst.RL achieved 2nd place in a computationally expensive RL challenge.
AE-DDPG improves DDPG for continuous control in complex environments with less data and time.
problem Data insufficiency and training inefficiency in DDPG for complex environments.
method Asynchronous episodic data collection, modified experience replay, and new action noise.
result AE-DDPG achieves higher rewards and less training time in complex environments.
SLM Lab is a framework for reproducible RL research with modular algorithms.
problem Reproducibility in deep reinforcement learning.
method Modular software framework for RL algorithms, synchronous/asynchronous execution, hyperparameter search, result analysis.
result Comprehensive benchmark and novel RL algorithms (e.g., discrete-AC variant, hybrid training method).
TorchBeast simplifies RL research in PyTorch.
problem Conducting scalable RL research with minimal programming knowledge.
method Pure-Python and C++ implementation for parallel training; OpenAI Gym interface.
result Performs on-par with IMPALA on Atari.
New method for asynchronous stochastic approximation converges in reinforcement learning.
problem Finding solutions to equations with noisy measurements in reinforcement learning.
method Batch Asynchronous Stochastic Approximation (BASA) with conditions for convergence and rate of convergence.
result Sufficient conditions for convergence and rate of convergence of BASA.
RL agent learns to place limit orders for trading signals in financial markets.
problem Training an RL agent to execute trading signals in limit order book markets.
method Deep Duelling Double Q-learning with APEX architecture, using synthetic alpha signals.
result RL agent outperforms heuristic trading strategies in inventory management and order placing.
Asynchronous framework speeds up model-based RL to real-time.
problem Real-time learning on real robots with model-based RL.
method Asynchronous framework for model-based reinforcement learning.
result Reduced run time to data collection time, improved sample complexity.
catalyst.RL simplifies RL research, enabling reproducibility and efficiency.
problem Difficulty in reproducing and comparing RL algorithms.
method Open-source framework with distributed training, flexible configurations, and efficient RL algorithms.
result Demonstrated effectiveness on AI for Prosthetics Challenge, achieving 3rd place.
We present an approach for reconfiguration of dynamic visual sensor networks with deep reinforcement learning (RL). Our RL agent uses a modified asynchronous advantage actor-critic framework and the recently proposed Relational Network module at the foundation of its network architecture. To address the issue of sample…
Enhances RL in partially observable, noisy environments by uncovering causal states.
problem Making decisions based on incomplete and noisy observations in partially observable Markov decision processes (P2OMDPs). method Causal State Representation under Asynchronous Diffusion Model (CaDiff) framework, incorporating a novel asynchronous diffusion model (ADM) and a new bisimulation metric.
result Enhances returns by at least 14.18% compared to baselines on Roboschool tasks.
AR-A3C improves A3C's robustness to noisy environments.
problem Neural networks are vulnerable to noise from unexpected sources in reinforcement learning.
method Introduced an adversarial agent to make the learning process more robust.
result AR-A3C outperforms A3C in both clean and noisy environments.
D2D-SPL uses discrete states and a classifier to train RL faster.
problem Training neural networks in RL due to correlated samples.
method Discretizes state space, uses actor-critic, selects input/target pairs, trains classifier.
result Trains faster than state-of-the-art methods.
Federated Q-learning achieves linear speedup with heterogeneity, improving sample complexity.
problem Collaborative learning in distributed RL settings with limited data sharing.
method Analyzes synchronous and asynchronous federated Q-learning, proposes importance averaging.
result Achieves linear speedup with heterogeneity, robust to local trajectory heterogeneity.
Improved sample efficiency in reinforcement learning with action guidance.
problem Sample inefficiency in deep reinforcement learning with sparse, delayed, and deceptive rewards.
method Integrating a non-expert demonstrator (e.g., MCTS) into asynchronous distributed deep reinforcement learning.
result Our methods learn faster and converge to better policies on a game.
IMPACT improves RL training speed without sacrificing sample efficiency.
problem Limited sample efficiency in scalable RL architectures.
method Proposes IMPACT, extending IMPALA with target networks, circular buffers, and truncated importance sampling.
result IMPACT achieves higher rewards and significantly reduces training time compared to IMPALA.
Deep reinforcement learning (deep RL) has achieved superior performance in complex sequential tasks by using deep neural networks as function approximators to learn directly from raw input images. However, learning directly from raw images is data inefficient. The agent must learn feature representation of complex stat…
Deep RL multi-task learning outperforms single-task learning on new tasks.
problem Improving performance on new tasks in multi-task reinforcement learning.
method Investigation of multi-task reinforcement learning algorithms with and without Elastic Weight Consolidation (EWC).
result Multi-task reinforcement learning algorithms outperform single-task learning on new tasks.
GraSP-RL uses graph neural networks to improve job shop scheduling.
problem Capturing machine-unit-job sequence relationships and managing state space growth.
method Graph neural networks for feature extraction, reinforcement learning for decision-making, decentralized optimization.
result GraSP-RL outperforms existing methods in minimizing makespan for complex production environments.
Ringmaster ASGD improves Asynchronous SGD's efficiency under varying worker times.
problem Suboptimal performance of Asynchronous SGD under heterogeneous worker computation times.
method Ringmaster ASGD, a novel Asynchronous SGD method with optimal time complexity.
result Ringmaster ASGD achieves optimal time complexity under arbitrary worker heterogeneity.
Paper proposes Terminal Prediction to improve deep RL performance.
problem Sample inefficiency and convergence to locally optimal policies in deep reinforcement learning.
method Introduces a self-supervised auxiliary task, Terminal Prediction, to help representation learning.
result A3C-TP outperforms standard A3C in most domains and provides significant improvement in Pommerman.
Zeno++ improves robustness of asynchronous SGD in fully asynchronous settings.
problem Byzantine failures in fully asynchronous SGD.
method Estimates descent of loss after applying candidate gradient.
result Proves convergence for non-convex problems under Byzantine failures.
ALMAB-DC optimizes expensive black-box experiments using active learning and distributed computing.
problem Efficiently optimizing expensive, gradient-free objectives in computational statistics and machine learning.
method Combines active learning, multi-armed bandits, and distributed asynchronous computing.
result Achieves lower simple regret and superior performance in various tasks compared to non-ALMAB baselines.
AEGiS optimizes expensive function evaluations asynchronously.
problem Optimizing expensive black-box functions efficiently.
method Asynchronous ε-Greedy Bayesian Optimisation combining greedy search and Thompson sampling.
result AEGiS outperforms existing asynchronous BO methods.
Secure aggregation for buffered asynchronous federated learning without TEEs.
problem Privacy and convergence in buffered asynchronous federated learning.
method Developed a new protocol (BASecAgg) that ensures privacy without TEEs by carefully designing masks.
result BASecAgg achieves similar convergence guarantees as FedBuff without TEEs.
Standard acquisition functions are sufficient for asynchronous Bayesian optimization.
problem Redundant and repeated queries in asynchronous Bayesian optimization.
method Conceptual analysis and theoretical guarantees of standard acquisitions.
result Standard acquisition functions achieve theoretical guarantees equivalent to Thompson sampling in asynchronous settings.
Paper analyzes convergence and speedup of asynchronous parallel SGD.
problem Achieving good convergence and linear speedup in asynchronous parallel SGD.
method Second-order convergence analysis of APSGD with consistent read near strictly saddle points.
result Theoretical guarantee for using at most O(K1/3M−1/3) workers for good convergence and linear speedup. Batch Bayesian optimisation (BO) has been successfully applied to hyperparameter tuning using parallel computing, but it is wasteful of resources: workers that complete jobs ahead of others are left idle. We address this problem by developing an approach, Penalising Locally for Asynchronous Bayesian Optimisation on k…
Asynchronous method for hyperparameter and neural architecture search.
problem Efficiently searching for optimal hyperparameters and neural architectures.
method Model-based, asynchronous multi-fidelity method combining Hyperband and Gaussian process-based Bayesian optimization.
result Substantial speed-ups over current state-of-the-art methods on various benchmarks.
Unified coagent network theory with execution paths.
problem Hierarchical reinforcement learning exploration-exploitation trade-off.
method Formalized coagent networks, introduced execution paths, revisited coagent network theory.
result Shorter proof of policy gradient theorem, generalizable to asynchronous coagents.
This work tackles resource allocation in asynchronous and stochastic systems.
problem Distributed resource allocation in asynchronous and stochastic settings.
method Approximate stochastic primal-dual approach with asynchronous updates.
result The Asynchronous stochastic Primal-Dual (Asyn-PD) algorithm converges to the saddle point solution at a rate of O(1/t). Advances in asynchronous optimization methods for machine learning.
problem Efficiently solving large-scale optimization problems in machine learning.
method Asynchronous parallel and distributed optimization methods, accounting for information delays.
result Degree of asynchrony impacts convergence rates in stochastic optimization methods.
New Fourier transform method handles missing data and asynchronous observations.
problem Volatility inference with missing or asynchronous data.
method Spectral framework using Fourier transforms.
result Consistent volatility functional estimation with limit distributions.
New rules found to maintain neural network performance in asynchronous training.
problem Asynchronous training leads to degradation in generalization.
method Examined dynamical stability, derived rules for learning rate adjustment.
result Learning rate should be inversely proportional to delay for high delay values.
DANA mitigates gradient staleness in asynchronous distributed SGD with momentum.
problem Gradient staleness in asynchronous distributed SGD with momentum.
method DANA: a novel technique for asynchronous distributed SGD with momentum that computes the gradient on an estimated future position of the model's parameters.
result DANA fully incorporates momentum in asynchronous training with almost no ramifications to final accuracy.
As datasets continue to increase in size and multi-core computer architectures are developed, asynchronous parallel optimization algorithms become more and more essential to the field of Machine Learning. Unfortunately, conducting the theoretical analysis asynchronous methods is difficult, notably due to the introducti…
POAP and pySOT improve surrogate optimization of expensive functions.
problem Optimizing expensive functions with concurrent evaluations.
method Event-driven asynchronous framework for optimization strategies.
result Asynchronous computation offers significant speed-up advantages.
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.
Sparsification improves convergence in asynchronous distributed SGD, even in the presence of staleness.
problem Staleness in asynchronous distributed SGD.
method Applied sparsification to reduce communication overheads in distributed asynchronous settings.
result The ergodic convergence rate of sparsified asynchronous SGD matches that of vanilla SGD, $\mathcal{O} \left( 1/\sqrt{T}
ight)$, even in the presence of staleness.
The asymptotic pseudo-trajectory approach to stochastic approximation of Benaim, Hofbauer and Sorin is extended for asynchronous stochastic approximations with a set-valued mean field. The asynchronicity of the process is incorporated into the mean field to produce convergence results which remain similar to those of a…
We show that asymptotically, completely asynchronous stochastic gradient procedures achieve optimal (even to constant factors) convergence rates for the solution of convex optimization problems under nearly the same conditions required for asymptotic optimality of standard stochastic gradient procedures. Roughly, the n…
Distributed asynchronous SGD has become widely used for deep learning in large-scale systems, but remains notorious for its instability when increasing the number of workers. In this work, we study the dynamics of distributed asynchronous SGD under the lens of Lagrangian mechanics. Using this description, we introduce …
Paper proposes a new RL approach combining IL and RL methods to improve decision-making.
problem Challenges in RL with large state and action spaces, and difficulty in reward determination.
method Combines Imitation Learning and RL methods (SARSA and A3C) to learn sequential decision-making policies.
result Significantly decreases human effort and exploration time in learning decision-making policies.
Activities in reinforcement learning (RL) revolve around learning the Markov decision process (MDP) model, in particular, the following parameters: state values, V; state-action values, Q; and policy, pi. These parameters are commonly implemented as an array. Scaling up the problem means scaling up the size of the arra…
Unified analysis of asynchronous-SGD algorithms for distributed learning.
problem Analyzing asynchronous-SGD in heterogeneous settings with varying speeds and data distributions.
method Unified convergence theory for non-convex smooth functions, including pure asynchronous SGD and its modifications.
result Unified convergence rates for various asynchronous algorithms, including novel methods.
Paper proves CLTs for Q-learning with asynchronous updates.
problem Establishing convergence rates for Q-learning algorithms.
method Polyak-Ruppert averaging, non-asymptotic and functional CLTs.
result Convergence rates in Wasserstein distance for Q-learning.
Accelerates optimization in asynchronous systems with sparse updates.
problem Optimizing finite-sum objectives in asynchronous lock-free environments.
method New accelerated SVRG variant with sparse updates.
result Achieves optimal incremental gradient complexity.