Paper analyzes convergence of continual learning with adaptive methods.
problem Preventing catastrophic forgetting in sequential learning tasks.
method Adaptive method for nonconvex continual learning (NCCL) adjusts step sizes of previous and current tasks.
result Proposed adaptive method achieves same convergence rate as SGD when catastrophic forgetting is suppressed.
Continuity method proves existence of Mabuchi solitons on Fano manifolds.
problem Existence of Mabuchi solitons on extremal Fano manifolds.
method Continuity method applied to Mabuchi's generalized Kähler-Einstein metrics.
result Analytic proof of Mabuchi soliton existence without minimal model program.
EBMs improve continual learning without external memory or regularization.
problem Improving continual learning without external memory or regularization.
method Energy-Based Models with contrastive divergence training objective.
result EBMs outperform baseline methods on various benchmarks.
Study on curvature blow-up and convergence of continuity method on Hirzebruch surface.
problem Curvature blow-up and convergence of continuity method on Hirzebruch surface.
method Continuity method applied to generalised Hirzebruch surface, focusing on Gromov-Hausdorff convergence and scalar curvature estimates.
result A general solution to the continuity method either exists or all times, or the scalar curvature blows up.
New method uses continuous OT for fairness, outperforming discrete OT.
problem Fairness issues in machine learning models.
method Stochastic-gradient fairness method based on continuous optimal transport.
result Continuous OT method outperforms discrete OT when data is limited.
Develops DPG methods for continuous-time RL with deterministic policies.
problem High variance and slow convergence in stochastic policy RL methods.
method Derives continuous-time policy gradient formula and proposes CT-DDPG algorithm.
result CT-DDPG achieves superior stability and faster convergence in continuous-time RL.
Estimates curvature for long-time continuity method solutions.
problem Curvature estimates for long-time continuity method solutions.
method Adapting arguments from Kähler-Ricci flow to semi-ample canonical line bundles.
result Derives curvature bounds for product manifolds.
Paper proves method for calculating NML code length works for continuous models.
problem Uncertainty in calculating NML code length for continuous models.
method Introduced a novel decomposition approach based on the coarea formula to prove correctness for continuous cases.
result Method accurately calculates NML code length for continuous models.
We study finite-time collapsing limits of the continuity method. When the continuity method starting from a rational initial Kähler metric on a projective manifold encounters a finite-time volume collapsing, this projective manifold admits a Fano fibration over a lower dimensional base. In this case, we prove the conti…
Study develops method for estimating causal effects in continuous variables.
problem Lack of methods for estimating causal effects in continuous variables.
method Develops a method independent of data generating models for continuous variable interventions.
result Preserves identifiability of data and applies to any generating models.
Proposes first method for continuously indexed domain adaptation.
problem Challenges of transferring knowledge between continuously indexed domains.
method Combines adversarial adaptation with a novel discriminator.
result Outperforms state-of-the-art methods on synthetic and real-world datasets.
New method tackles anomaly detection in video surveillance using continual learning.
problem Challenges in continual learning for high-dimensional applications like video surveillance.
method Transfer learning and continual learning for online anomaly detection.
result Significantly reduces training complexity and continual learning from recent data.
Bilevel Continual Learning improves continual learning by transferring knowledge effectively.
problem Catastrophic forgetting and poor generalization in continual learning.
method Bilevel optimization and dual memory management strategies.
result BCL achieves effective knowledge transfer and alleviates catastrophic forgetting.
Continuous reinforcement learning such as DDPG and A3C are widely used in robot control and autonomous driving. However, both methods have theoretical weaknesses. While DDPG cannot control noises in the control process, A3C does not satisfy the continuity conditions under the Gaussian policy. To address these concerns,…
This paper presents a method to improve continual learning stability and plasticity.
problem Balancing learning stability and plasticity in deep learning.
method Batch-level Experience Replay with Review approach.
result Achieved 1st place in all three scenarios of the CVPR 2020 CLVision challenge.
Develops deep jump learning for continuous treatment OPE.
problem Estimating mean outcomes under new treatment rules using historical data from different rules.
method Adaptive deep discretization of continuous treatment space using deep learning and multi-scale change point detection.
result Validated method through theoretical results, simulations, and real application to Warfarin Dosing.
CPR adds entropy maximization to improve continual learning methods.
problem Catastrophic forgetting in continual learning.
method Classifier-Projection Regularization (CPR) adds an entropy maximization term to existing regularization methods.
result CPR improves accuracy and plasticity in continual learning methods.
Continuous-time MBRL framework tackles control systems with Bayesian ODEs.
problem Discretization of continuous-time systems in MBRL.
method Novel actor-critic method with Bayesian ODEs for state inference.
result Model robust against irregular and noisy data, sample-efficient, solves challenging control problems.
Efficiently scales continuous kernels with sparse Fourier domain learning.
problem High computational and memory demands, spectral bias in continuous kernels.
method Sparse learning in the Fourier domain.
result Efficient scaling of continuous kernels, reduced computational and memory requirements, mitigated spectral bias.
Application of discrete-time survival methods for continuous-time survival prediction is considered. For this purpose, a scheme for discretization of continuous-time data is proposed by considering the quantiles of the estimated event-time distribution, and, for smaller data sets, it is found to be preferable over the …
Study on continuity of solutions for complex Monge-Ampère equations with movable singularities.
problem Continuity of solutions with prescribed singularities for complex Monge-Ampère equations.
method Strong continuity methods with movable singularities, including Kähler-Einstein metrics.
result Sufficient conditions for strong continuity of solutions and openness results for Fano type equations.
In this paper we study a special case of the completion of cusp Kähler-Einstein metric on the regular part of varieties by taking the continuity method proposed by La Nave and Tian. The differential geometric and algebro-geometric properties of the noncollapsing limit in the continuity method with cusp singularities wi…
Paper proposes SDRL to improve continual learning with less computational cost.
problem Catastrophic forgetting in continual learning.
method SDRL method that refines gradients from memorized samples to reduce gradient diversity.
result SDRL shows better performance than state-of-the-art methods on multiple benchmark tasks.
Fluid approximations have seen great success in approximating the macro-scale behaviour of Markov systems with a large number of discrete states. However, these methods rely on the continuous-time Markov chain (CTMC) having a particular population structure which suggests a natural continuous state-space endowed with a…
In order to mimic the human ability of continual acquisition and transfer of knowledge across various tasks, a learning system needs the capability for continual learning, effectively utilizing the previously acquired skills. As such, the key challenge is to transfer and generalize the knowledge learned from one task t…
Research shows continual learning challenges in confounded datasets.
problem Challenges in mitigating confounders in continual learning settings.
method Formal description of continual confounders, construction of ConCon dataset.
result Standard continual learning methods fail to ignore confounders.
EWC helps prevent forgetting in neural networks by adjusting weights dynamically.
problem Preventing forgetting in neural networks during training.
method EWC adjusts weights dynamically to prevent forgetting.
result EWC effectively prevents catastrophic forgetting in neural networks.
New methods for inferring, predicting, and estimating continuous-time, discrete-event processes.
problem Inferring, predicting, and estimating entropy rate of continuous-time, discrete-event processes.
method Bayesian structural inference extended with neural networks.
result Methods are competitive for prediction and entropy-rate estimation with state-of-the-art.
Efficient method for learning continuous exponential families beyond Gaussian.
problem Learning continuous exponential families with unbounded support.
method Interaction Screening approach for scalable learning of continuous graphical models.
result Our estimator maintains similar accuracy and sample complexity scalings compared to alternative approaches, while improving run-time.
We propose a novel continual learning method called Residual Continual Learning (ResCL). Our method can prevent the catastrophic forgetting phenomenon in sequential learning of multiple tasks, without any source task information except the original network. ResCL reparameterizes network parameters by linearly combining…
New method uses dynamic programming for meta continual learning.
problem Challenges of generalization and catastrophic forgetting in sequential learning.
method Developed a theoretical framework using dynamic programming for meta continual learning.
result Theoretical and practical method achieves better accuracy than existing methods.
Derives effective continuous dynamics for adaptive SGD methods.
problem Analyzing noise in adaptive SGD methods.
method Stochastic modified equations framework and Malladi's scaling rules.
result Sampling-induced noise in SGD limits to independent Brownian motions.
Study extends continuity equation for Gauduchon metrics.
problem Continuity equation for Gauduchon metrics.
method Solution to Gauduchon conjecture by Székelyhidi, Tosatti, and Weinkove.
result Extended interval of maximal existence for continuity equation.
Neural models learn continuous-time Markov chain transition rates from data.
problem Learning transition rates for complex stochastic systems.
method Neural networks to model nonlinear transition rates from observed data.
result Neural models outperform traditional methods in accuracy.
New method detects symmetries beyond affine transformations.
problem Current methods limit symmetry detection to affine transformations.
method Framework for discovering continuous symmetry beyond affine transformations.
result Method is competitive for large sample sizes and superior for small sample sizes.
Paper proposes a new method to select memory data for online class-incremental learning.
problem Selecting which buffered images to replay for online class-incremental learning.
method Adversarial Shapley value scoring method to preserve latent decision boundaries.
result Proposed ASER method provides competitive or improved performance compared to state-of-the-art methods.
AGS-CL selectively updates penalties based on node importance for continual learning.
problem Catastrophic forgetting in continual learning.
method Adaptive Group Sparsity (AGS) with proximal gradient descent.
result Significantly outperforms baselines on various continual learning benchmarks.
Hybrid RL method optimizes trading by balancing continuous and discrete actions.
problem Optimal execution in algorithmic trading with continuous-discrete action space.
method Combines continuous and discrete RL agents for better trading decisions.
result Significantly outperforms existing methods in trading efficiency and stability.
New method solves continuous time mean-variance model for consistent investment strategy.
problem Time-consistent optimal strategy for continuous time mean-variance model.
method Developed a new Bellman principle method.
result Obtained a time-consistent dynamic optimal strategy.
A new method for Gaussian Processes handles mixed continuous and categorical inputs.
problem Modeling cross-correlations between continuous and categorical data.
method Low-Rank Correlation (LRC) method for Gaussian Processes with flexible rank approximation.
result LRC outperforms existing methods in estimating cross-correlations and predicting response surfaces.
We propose a novel score-based approach to learning a directed acyclic graph (DAG) from observational data. We adapt a recently proposed continuous constrained optimization formulation to allow for nonlinear relationships between variables using neural networks. This extension allows to model complex interactions while…
Efficient deep policy gradient method for continuous-time control problems.
problem Optimal control in continuous time with fine time discretization.
method Multi-scale deep policy gradient method with varying time discretization.
result Targeted efficiency in computational resources achieved through multi-scale approach.
New algorithm estimates semi-continuous data density using entropy maximization.
problem Estimating density functions for semi-continuous data.
method Maximum entropy principle, requiring only constraint function samples.
result Estimate has significantly less bias compared to existing methods.
Paper introduces new estimator for continuous treatment effects.
problem Estimating the average dose-response function of continuous treatments.
method Utilizes ADML and DML tools, with a novel debiasing method.
result Proves asymptotic normality and shows good performance in simulations.
This paper develops variational continual learning (VCL), a simple but general framework for continual learning that fuses online variational inference (VI) and recent advances in Monte Carlo VI for neural networks. The framework can successfully train both deep discriminative models and deep generative models in compl…
Paper proposes copula-based models for analyzing multivariate zero-inflated continuous data.
problem Challenges in analyzing multivariate zero-inflated continuous data with mixed discreteness and continuity.
method Proposes two copula-based density estimation models and rectified Gaussian copula.
result Demonstrates superior performance compared to conventional methods.
Estimating causal models from observational data is a crucial task in data analysis. For continuous-valued data, Shimizu et al. have proposed a linear acyclic non-Gaussian model to understand the data generating process, and have shown that their model is identifiable when the number of data is sufficiently large. Howe…
Many scientific questions require estimating the effects of continuous treatments. Outcome modeling and weighted regression based on the generalized propensity score are the most commonly used methods to evaluate continuous effects. However, these techniques may be sensitive to model misspecification, extreme weights o…