Model learns continuously from text without forgetting.
problem Catastrophic forgetting in lifelong language learning.
method Episodic memory with sparse experience replay and local adaptation.
result Model can continuously learn from new datasets.
On-device research index
arXiv research
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.
169,291 papers · 148 categories
Trend · papers per month
48 results for “catastrophic episodes”
A novel approach stores encoded images as centroids and covariance matrices to improve classification accuracy with less memory.
problem Catastrophic forgetting and memory limitations in continual learning.
method Trains autoencoders with Neural Style Transfer to encode images, replay encoded episodes to avoid forgetting, and use centroids and covariance matrices for pseudo-images when memory is full.
result Increases classification accuracy by 13-17% over state-of-the-art methods on benchmark datasets, while requiring 78% less storage space.
Optimal CL requires perfect memory and is NP-hard.
problem Designing CL algorithms that perform reliably and avoid catastrophic forgetting.
method Theoretical approach to derive computational properties of optimal CL algorithms.
result Optimal CL algorithms generally solve an NP-hard problem and require perfect memory.
FSR prevents deep nets from forgetting old tasks.
problem Catastrophic forgetting in deep neural networks.
method Logit matching on selected samples from old tasks.
result FSR outperforms previous methods in knowledge retention.
SALeRA controls SGD learning rate to learn as fast as possible but not faster.
problem Catastrophic learning episodes in SGD training of deep neural networks.
method SALeRA uses two statistical tests: one for speeding up and one for detecting and halting catastrophic episodes.
result SALeRA learns as fast as possible but not faster, as demonstrated by experiments on standard benchmarks.
Proposes a CL technique to improve accuracy and reduce forgetting.
problem Sequential task learners struggle with forgetting information from previous tasks.
method Extracts modular parts of neural networks and estimates task relatedness.
result Remarkable performance gain in robustness to forgetting for EWC and GEM methods.
Unified approach for lifelong learning in recurrent neural networks.
problem Catastrophic forgetting and capacity saturation in lifelong learning.
method Proposed a curriculum-based benchmark and unified model combining Gradient Episodic Memory and Net2Net.
result Unified model performs better than constituent models in lifelong learning setting.
New method improves continual learning by anchoring past knowledge.
problem Catastrophic forgetting in continual learning.
method Bilevel optimization to update current task knowledge while keeping past task predictions.
result Improves accuracy and forgetting metrics compared to experience replay.
New schemes improve lifelong learning by balancing old and new tasks.
problem Catastrophic forgetting in deep neural networks when learning multiple tasks.
method Unified optimization perspective of episodic memory based approaches, introducing MEGA-I and MEGA-II schemes.
result Significant improvement in lifelong learning benchmarks, reducing error by up to 18%.
A new method learns from multi-modal sequences with external memory.
problem Learning new modes in a dynamic environment without prior knowledge.
method Maintains a neural episodic memory with a Dirichlet Process prior to store mode descriptors and transfers knowledge through retrieval.
result Performs continual learning favorably compared to mainstream approaches.
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.
La-MAML improves fast online continual learning with a look-ahead approach.
problem Fast online continual learning with limited model capacity.
method Optimisation-based meta-learning with look-ahead and episodic memory.
result Superior performance on visual classification benchmarks.
Graph-based rehearsal improves continual learning performance.
problem Catastrophic forgetting in continual learning models.
method Augmenting a memory array with a learnable random graph to capture pairwise similarities.
result Our model consistently outperforms baselines on task-free continual learning benchmarks.
Efficiently stores and retrieves past states for faster learning in reinforcement learning.
problem Data inefficiency and memory limitations in reinforcement learning.
method Dynamic online k-means for state clustering and prioritization.
result Dynamic online k-means improves performance with smaller memory sizes.
Sparse routing networks with co-training prevent catastrophic forgetting in continual learning.
problem Catastrophic forgetting in neural networks trained on a sequence of tasks.
method Sparse routing networks with co-training to minimize interference between dissimilar tasks.
result Sparse routing networks with co-training outperform densely connected networks on benchmarks.
Safe reinforcement learning tackles safety constraints with linear approximations.
problem Ensuring safety in reinforcement learning without violating constraints.
method Modeling safety as a linear cost function, developing SLUCB-QVI and RSLUCB-QVI algorithms for MDPs with linear function approximation.
result Achieved a nearly optimal regret bound for safe reinforcement learning, matching state-of-the-art unsafe algorithms.
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.
New method improves ABI for sequential data, reducing forgetting and improving accuracy.
problem Performance degradation of ABI under model misspecification and distribution shifts.
method Decouples simulation-based pre-training from unsupervised SC fine-tuning, using memory buffer and elastic weight consolidation.
result Significant mitigation of forgetting and improved posterior estimates compared to standard simulation-based training.
A method for continual learning using world models in reinforcement learning.
problem Catastrophic forgetting in lifelong learning with neural networks.
method Interleaving internally generated episodes of past experiences (pseudo-rehearsal) with external environment's observations.
result Consistent reduction in temporal prediction loss compared to non-interleaved learning.
A model retains learned knowledge for longer by adding a plastic component to neural networks.
problem Catastrophic forgetting in neural networks when learning new tasks.
method Differentiable Hebbian Consolidation model with a DHP Softmax layer.
result Reduces forgetting in benchmarks like Permuted MNIST and Vision Datasets Mixture.
Paper tackles forgetting in neural networks, proposing solutions.
problem Catastrophic forgetting in artificial neural networks.
method Simple model and reinforcement learning applications.
result Proposes solutions to prevent forgetting in neural networks.
A new method helps deep learning systems adapt to changing conditions.
problem Deep learning systems struggle with environmental drifts and long healing cycles.
method Intentional forgetting integrated into continual learning to overcome issues.
result Dr. DRL reduces healing time and fine-tuning episodes by 18.74% and 17.72% respectively.
The study values a new type of insurance-linked security called CocoCat bonds.
problem Valuing a new type of insurance-linked security called contingent convertible catastrophe bonds.
method Formalized design, derived analytical valuation formulae, used time-inhomogeneous compound Poisson process for natural catastrophe losses, and applied exponential change of measure and Girsanov-like transformation.
result CocoCat bond prices are most sensitive to interest rates, conversion fractions, and trigger levels.
A neural network approach to learn Cusp Catastrophe dynamics.
problem Complex behavior and non-convex parameter space in Cusp Catastrophe models.
method Training a deep neural network to learn dynamics without solving generating parameters.
result Demonstrated a neural network approach for the first time in Cusp Catastrophe models.
The paper values reinsurance contracts for dynamic catastrophe claims without arbitrage.
problem Valuation of reinsurance contracts for dynamic catastrophe claims without arbitrage.
method Compound dynamic contagion process, Esscher transform, Monte Carlo simulation.
result Arbitrage-free premiums for catastrophe stop-loss reinsurance contracts.
A random forest based approach for predicting spreads in the primary catastrophe bond marketq-fin.PR
Random forest predicts catastrophe bond spreads with 93% accuracy.
problem Predicting spreads in the primary catastrophe bond market.
method Random forest approach using all information in offering circulars.
result Random forest explains 93% of spread variability, significantly better than linear regression (47%).
Adam optimizer leads to more forgetting in neural networks.
problem Understanding and quantifying catastrophic forgetting in neural networks.
method Comparative analysis of various optimization algorithms and metrics in different learning scenarios.
result Adam optimizer causes more forgetting compared to classical algorithms like SGD.
Bayesian online meta-learning framework tackles catastrophic forgetting in few-shot classification.
problem Catastrophic forgetting in few-shot classification problems.
method Bayesian online learning, meta-learning, Laplace approximation, variational inference.
result Framework effectively achieves goal of overcoming catastrophic forgetting in few-shot classification.
Unified Bayesian framework for CAT bond pricing.
problem Uncertainty in catastrophe occurrences and interest rates in CAT bond markets.
method Bayesian framework based on uncertainty quantification of catastrophes and interest rates.
result Unified asset pricing approach with informative expected risk premia.
Paper introduces Auto DeepVis to explain catastrophic forgetting in continual learning.
problem Catastrophic forgetting in continual learning of deep neural networks.
method Auto DeepVis and critical freezing techniques to address catastrophic forgetting.
result Critical freezing outperforms other methods on both past and future tasks.
Detecting early signs of mood episodes in bipolar disorder patients.
problem Early identification of mood episodes in bipolar disorder patients for timely treatment.
method Signature-based model derived from stochastic analysis applied to real-time mood data.
result The signature method can identify the onset of mood episodes in bipolar disorder patients.
Optimizes diversification in catastrophe risk pooling using asymptotic analysis.
problem Maximizing diversification benefit from catastrophic events in insurance pools.
method Asymptotic analysis to solve high-dimensional optimization problem.
result Derives an asymptotically optimal pool that approximates practical optimal pool.
Study helps identify which ANN parameters cause forgetting.
problem Catastrophic forgetting in neural networks.
method Determines individual parameter contributions to forgetting.
result Identifies specific ANN parameters causing forgetting.
Intrinsic fear prevents catastrophic states in reinforcement learning.
problem Catastrophic forgetting in reinforcement learning.
method Intrinsic fear (IF) is a learned reward shaping that penalizes the Q-learning objective based on the probability of imminent catastrophe.
result Intrinsic fear models prevent periodic catastrophes in reinforcement learning agents.
EMDQN uses episodic memory to improve RL efficiency.
problem Sample inefficiency of deep RL algorithms.
method Leverages episodic memory to supervise training.
result Significantly reduces interaction rounds for state-of-the-art performance.
Study preference-based reinforcement learning in episodic kernel MDPs.
problem Learning from episodic human preferences in reinforcement learning.
method Developed preference-based value estimation and confidence sets for kernel-based MDPs.
result Proved high-probability regret bounds that converge to optimal policy value.
Algorithm tackles constrained reinforcement learning with concave-convex and knapsack constraints.
problem Constrained episodic reinforcement learning with concave rewards and convex constraints.
method Modular analysis with strong theoretical guarantees for concave-convex and knapsack settings.
result Significantly outperforms existing approaches in constrained episodic environments.
Algorithm improves multi-armed bandit performance by transferring reward samples.
problem Sequential multi-armed bandit problem with changing reward distributions.
method UCB algorithm with reward sample transfer.
result Significant improvement in cumulative regret over standard UCB.
Paper tackles catastrophic overfitting in single-step adversarial training.
problem Catastrophic overfitting leads to sudden drop in robust accuracy.
method Proposes a method to prevent overfitting by using all adversarial examples.
result Demonstrates prevention of catastrophic overfitting and improves robustness.
BerlinUCB learns from episodic rewards in nonstationary contexts.
problem Online learning with episodic rewards in nonstationary environments.
method BerlinUCB integrates clustering for self-supervision.
result BerlinUCB outperforms standard contextual bandits in various scenarios.
New methods improve insurance data quality for catastrophic events.
problem Improving precision and size of insurance data for catastrophic events.
method Bootstrap, bootknife, and GAN algorithms.
result Compared MSE and MAE of simulated outputs, direct algorithm for fuzzy expert opinion.
EBU improves deep RL with episodic backward updates, achieving human-level performance with less data.
problem Efficiently learn from sparse and delayed rewards in reinforcement learning.
method Episodic Backward Update (EBU) algorithm that recursively propagates value updates.
result EBU achieves human-level performance in Atari games with significantly fewer samples.
This paper develops a two-step estimation methodology, which allows us to apply catastrophe theory to stock market returns with time-varying volatility and model stock market crashes. Utilizing high frequency data, we estimate the daily realized volatility from the returns in the first step and use stochastic cusp cata…
Study analyzes catastrophic forgetting in continual learning using teacher-student networks.
problem Catastrophic forgetting in continuously learning systems.
method Teacher-student learning framework, similarity of input distributions and target functions.
result Network can avoid catastrophic forgetting with small input distribution similarity and large target function similarity.
The paper introduces CoCoCat bonds for multi-region natural catastrophes, accounting for complex dependencies.
problem Valuation of multi-region contingent convertible bonds under complex dependencies.
method Developed a model accounting for inter-regional dependencies using change-of-measure techniques.
result Significant impact of inter-regional dependencies on CoCoCat bond pricing.
Improved regret bound for online learning in unknown MDPs.
problem Online learning in unknown episodic MDPs with changing loss functions.
method Adapts adversarial MDP model to convex performance criteria using entropic regularization.
result Achieved regret bound.
The paper examines GANs' forgetting and mode collapse, showing how they relate and impact training.
problem Catastrophic forgetting and mode collapse in GANs during continual learning.
method Investigates the continual learning nature of GANs, analyzing discriminator's output landscapes and convergence.
result Catastrophic forgetting and mode collapse are interrelated and prevent GANs from converging.
Algorithm reduces episode count for CMDPs with constraints.
problem Online decision-making with constraints in episodic CMDPs.
method Optimistic planning using linear programming for PAC guarantee.
result Probably approximately correct (PAC) guarantee on episode count.