The paper analyzes errors in terminating reinforcement learning prematurely.
problem Premature termination of reinforcement learning leads to significant errors.
method Combinatorial analysis and simulation experiments.
result The error of premature termination can be quite high and cannot be eliminated.
A new BO termination criterion for HPO reduces optimization time without sacrificing test performance.
problem Determining an optimal budget for hyperparameter optimization.
method A new termination criterion based on the discrepancy between predictive and computable target performance.
result The proposed termination criterion achieves a better trade-off between test performance and optimization time.
Full-batch GD achieves generalization close to any stationary point with fewer assumptions.
problem Generalization and excess risk bounds for smooth losses, including non-Lipschitz and nonconvex cases.
method Path-dependent analysis of GD's generalization error, focusing on optimization error and stability.
result Generalization error is tightly bound in terms of optimization error and iteration count, bypassing common assumptions.
Energy-efficient detection of natural errors in deep networks.
problem Deep networks lack error detection capability without additional energy costs.
method Append RACs at hidden layers to detect natural errors with early classification termination.
result Early classification termination reduces energy consumption.
Social Security and other public policies can be viewed as a series of cash in and outflows that depend on parameters such as the age distribution of the population and the retirement age. Given forecasts of these parameters, policies can be designed to be financially stable, i.e., to terminate with a zero balance. If …
Selective inference framework for CART trees to control error rates and coverage.
problem Inference on CART trees does not control Type 1 error rates and coverage.
method Selective inference framework conditioning on tree estimation, efficient algorithms.
result Proposes tests and intervals for CART trees with selective error control.
Conditional diffusion models can approximate target distributions well with Gaussian-mixture reverse kernels.
problem Approximating target distributions in conditional diffusion models.
method Using finite Gaussian mixtures with ReLU-network logits as reverse kernels, reducing the problem to static conditional density approximation.
result The resulting neural reverse-kernel class is dense in conditional KL divergence under exact terminal matching.
Solves steering problem with continuous time, Hilbert-Schmidt cost, and matrix ODEs.
problem Fixed horizon linear quadratic covariance steering in continuous time with a specific terminal cost.
method Formulates necessary conditions as a coupled matrix ODE two-point boundary value problem, designs a matricial recursive algorithm, and proves convergence.
result Proposes and proves the convergence of a matricial recursive algorithm for solving the steering problem.
Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.
problem Understanding and optimizing deep learning training phases.
method Direct measurements on three deepnet architectures across seven datasets.
result Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.
D2SRM solves complex PDEs using deep learning.
problem High-dimensional, Hessian-dependent fully nonlinear parabolic PDEs.
method Single scalar space-time network generating derivative-consistent approximations trained through residuals and penalties.
result Well-posedness and convergence theory established for globally Lipschitz equations.
Proposes a new model to optimize investment plans with varying terminal times.
problem Improving the classical mean-variance model for continuous time investments.
method Uses stochastic optimal control and varying terminal time to determine optimal strategies.
result Optimal strategies and terminal times can be determined to minimize portfolio variance.
This paper improves non-asymptotic bounds for denoising diffusions, focusing on the Ornstein-Uhlenbeck process.
problem Improving non-asymptotic bounds for denoising diffusions, especially for the Ornstein-Uhlenbeck process.
method Explicit non-asymptotic bounds on forward diffusion error in total variation, considering multi-modal data distributions.
result The Ornstein-Uhlenbeck process cannot be significantly improved in terms of reducing terminal time T for multi-modal data distributions. A new algorithm discovers useful reinforcement learning options by focusing on termination compressibility.
problem Discovering useful reinforcement learning options.
method Proposes an algorithm that focuses on the termination condition, using a critic to learn the option transition model and optimize compressibility.
result The resulting options are non-trivial, meaningful, and useful for learning and planning.
Generative model prices basket options efficiently.
problem Real-time pricing of basket options with varying market inputs.
method Truncated path signatures and Mixture Density Networks (MDN) for learning the terminal density.
result The model produces small pricing errors and matches Monte Carlo simulations closely.
This paper explains CART random forests using stochastic control theory.
problem Understanding the inner workings of CART random forests.
method Developed a stochastic-control perspective on CART random forests, interpreting feature subsampling as a random feasible action set and the split rule as a policy.
result Established that the CART policy is locally stabilizing but globally suboptimal for the forest objective.
Study bounds for prices of European and American options with optional termination.
problem Bounding prices of options with potential termination.
method Duality results linking upper prices of vulnerable options to American options with constrained exercise times.
result Linking upper prices of vulnerable options to American options and game options.
The paper finds optimal threshold strategies for insurance companies with a positive terminal value at creeping ruin.
problem Optimizing dividend payments in an insurance company's surplus process with a positive terminal value at creeping ruin.
method Using fluctuation theory, the paper derives explicit formulas for the objective function and shows the optimality of threshold strategies.
result Threshold strategies are optimal for the dividend optimization problem under certain conditions.
New method preserves distances in time series data.
problem Preserving distances in time series data under interpolation.
method Developed lines-preserving terminal embeddings.
result First dimension-free coresets for Fréchet distance clustering.
Fast, reliable, and error-bounded option pricing with neural networks
problem Fast, reliable, and error-bounded option pricing
method Mixture Density Network
result Out-of-sample CDF error of 1.4imes10−4 Proves finite step termination of Kähler-Einstein metric singularity formation.
problem Singularity formation of Kähler-Einstein metrics.
method Finite step termination of bubble trees for singularity formation.
result Finite step termination of Kähler-Einstein metric singularity formation proved in non-collapsing situation.
The study proves a key inequality for specific types of three-dimensional spaces.
problem Establishing a mathematical inequality for a specific class of three-dimensional spaces.
method Developed the orbifold version of the Bogomolov-Gieseker inequality for stable Q-sheaves on log terminal Kähler threefolds.
result Proved the Bogomolov-Gieseker inequality for log terminal Kähler threefolds.
New test for SGD in binary classification reduces computation time.
problem Determining optimal stopping for SGD in binary classification.
method Proposes a new, simple, computationally inexpensive termination criterion for SGD.
result Termination criterion reduces expected misclassification probability.
Employee stock options (ESOs) are American-style call options that can be terminated early due to employment shock. This paper studies an ESO valuation framework that accounts for job termination risk and jumps in the company stock price. Under general Lévy stock price dynamics, we show that a higher job termination ri…
We establish existence, uniqueness and regularity of solution results for a class of backward stochastic partial differential equations with singular terminal condition. The equation describes the value function of non-Markovian stochastic optimal control problem in which the terminal state of the controlled process is…
New method for computing terminal embeddings in sublinear time.
problem Efficiently computing terminal embeddings with sublinear time complexity.
method Developed a data structure to compute terminal embeddings in sublinear time.
result Achieved sublinear time computation of terminal embeddings.
We prove that the sum of the α-invariants of two different Kollár components of a Kawamata log terminal singularity is less than 1.
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.
Locally adaptive clustering for tree delineation.
problem Tree delineation from distance data.
method Locally adaptive hierarchical cluster termination.
result Multi-scale alternative to conventional termination criteria.
Is an option to early terminate a swap at its market value worth zero? At first sight it is, but in presence of counterparty risk it depends on the criteria used to determine such market value. In case of a single uncollateralised swap transaction under ISDA between two defaultable counterparties, the additional unilat…
New loss function improves classification for imbalanced and sensitive groups.
problem Optimizing metrics like balanced error and equal opportunity in imbalanced and sensitive classification.
method Developed a principled vector-scaling (VS) loss function that addresses multiplicative adjustments necessary at terminal training phase.
result The VS loss function improves minority class performance and generalizes to different types of imbalances.
Study optimal liquidation with multiple regimes using BSDEs with singular terminal values.
problem Optimal liquidation with regime switching in dark pools.
method Introduced a system of BSDEs with jumps and singular terminal values.
result Existence and uniqueness results for the BSDE system are obtained.
This paper establishes the existence of a unique nonnegative continuous viscosity solution to the HJB equation associated with a Markovian linear-quadratic control problems with singular terminal state constraint and possibly unbounded cost coefficients. The existence result is based on a novel comparison principle for…
New reward function improves GAIL performance in task-based environments.
problem Reward bias in adversarial imitation learning.
method Proposed a new reward function to overcome existing biases.
result New reward function outperforms existing methods in task-based environments.
We provide representations of solutions to terminal value problems of inhomogeneous Black-Scholes equations and studied such general properties as min-max estimates, gradient estimates, monotonicity and convexity of the solutions with respect to the stock price variable, which are important for financial security prici…
Develops a learning model predictive controller for competitive racing.
problem Lack of exploration in state space and complexity in obstacle avoidance.
method Explores state space through multiple initializations and develops a new method for convex terminal set selection.
result Yields a richer terminal safe set and maintains convexity.
Circular nets with spherical parameter lines have geometric properties related to Darboux cyclides and terminating Laplace sequences.
problem Discretizing surfaces with spherical curvature lines.
method Lie-geometric discretisation in terms of principal contact element nets.
result Circular nets with two families of spherical parameter lines are related to Darboux cyclides.
A framework solves parametric families of MFGs efficiently.
problem Efficiently solving MFG systems with varying initial distributions and terminal costs.
method Operator learning framework for parametric families of MFGs.
result Accurate approximation for cybersecurity and quadratic MFGs.
New framework values ESOs with multiple exercises and job termination risk.
problem Valuing ESOs with complex exercise patterns and job termination risk.
method Fourier transform, finite differences, and maturity randomization methods.
result Analytic formulae for ESO costs under various conditions.
TVM improves generative modeling by matching terminal velocities.
problem Creating high-fidelity one- and few-step generative models.
method TVM generalizes flow matching, modeling transitions between diffusion timesteps and regularizing terminal behavior.
result TVM achieves state-of-the-art FID scores with minimal architectural changes and fused attention kernel.
We apply the language of the groupoid approach to Lie pseudo-groups, and the classical Cartan-Kuranishi theorem, to prove that Cartan's equivalence method terminates at involution (or at complete reduction) for constant type problems.
The paper analyzes and proposes a new stopping criterion for recursive Bayesian classification.
problem Limitations of conventional stopping criteria in recursive Bayesian classification.
method Geometric interpretation of state posterior progression and analysis of conventional criteria.
result Proposes a new stopping criterion to overcome limitations of conventional methods.
In this paper, the `Approximate Message Passing' (AMP) algorithm, initially developed for compressed sensing of signals under i.i.d. Gaussian measurement matrices, has been extended to a multi-terminal setting (MAMP algorithm). It has been shown that similar to its single terminal counterpart, the behavior of MAMP algo…
A new algorithm LONR learns without terminal states or perfect recall.
problem Learning in settings without terminal states or perfect recall.
method Local No-Regret Learning (LONR) using Q-learning-like updates.
result LONR achieves last iterate convergence in challenging settings.
Paper connects neural network score approximation to reverse diffusion model distribution approximation.
problem Quantifying the relationship between neural network score approximation and the distribution generated by reverse diffusion models.
method Combines Hornik's universal approximation theorem, Girsanov's theorem, and data processing inequality.
result Neural network score approximation guarantees distribution approximation in reverse diffusion models.
Researchers find Kähler-Einstein metrics near isolated log terminal singularities.
problem Existence of Kähler-Einstein metrics with positive curvature near isolated log terminal singularities.
method Solving complex Monge-Ampère equations to analyze the existence of metrics.
result Existence of smooth solutions in subcritical regimes, with critical exponent expressed in terms of normalized volume.
We consider joint estimation of multiple graphical models arising from heterogeneous and high-dimensional observations. Unlike most previous approaches which assume that the cluster structure is given in advance, an appealing feature of our method is to learn cluster structure while estimating heterogeneous graphical m…
Optimal asset allocation strategy outperforms stochastic benchmark.
problem Achieving higher terminal wealth than a stochastic benchmark.
method Data-driven Neural Network optimization framework for dynamic asset allocation.
result Optimal adaptive strategy outperforms benchmark with higher median and right-skewed terminal wealth.
ETCNN uses neural networks to price American options accurately.
problem Accurately pricing American options with inequality constraints.
method ETCNN framework solving BSM equations with exact terminal condition.
result ETCNN achieves high accuracy and robustness across various scenarios.