New method fine-tunes discrete diffusion models for RLHF tasks.
problem Fine-tuning discrete diffusion models with policy gradient methods is challenging.
method Proposed SEPO algorithm for efficient fine-tuning over non-differentiable rewards.
result Numerical experiments show scalability and efficiency of SEPO.
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.
Proposes DAM for optimizing discrete generative models.
problem Challenges in optimizing discrete generative models.
method Discrete Adjoint Matching (DAM) for discrete state spaces.
result Demonstrates effectiveness on synthetic and mathematical reasoning tasks.
In this paper, we develop the continuous time dynamic topic model (cDTM). The cDTM is a dynamic topic model that uses Brownian motion to model the latent topics through a sequential collection of documents, where a "topic" is a pattern of word use that we expect to evolve over the course of the collection. We derive an…
DTM improves dLLM fine-tuning stability and performance.
problem Intractable sequence-level marginal likelihoods for masked diffusion models.
method Discrete Tilt Matching (DTM) recasts dLLM fine-tuning as state-level matching of local unmasking posteriors under reward tilting.
result DTM yields strong gains on Sudoku and Countdown while remaining competitive on MATH500 and GSM8K.
Non-autoregressive method speeds up protein folding prediction 23 times.
problem Generating protein sequences with higher order interactions.
method Discrete diffusion conditioned on 3D structure using ProteinMPNN.
result 23 times speed up in inference without performance loss.
In an earlier work we identified the types and numbers of static equilibrium points of solids arising from fine, equidistant n-discretrizations of smooth, convex surfaces. We showed that such discretizations carry equilibrium points on two scales: the local scale corresponds to the discretization, the global scale to…
Improved Markov models learn from their mistakes and adapt to problem complexity.
problem Limitations of standard masked discrete diffusion models in reasoning tasks.
method Learning a Markov transition kernel trained on its own outputs, allowing remasking and adaptation.
result Significant improvement in solving reasoning problems, especially Sudoku-Extreme and Countdown-4.
DIF extends NF with stochastic discrete latent variables for better density estimation.
problem Improving density estimation with discontinuities and fine details.
method Discretely indexed flows as an extension of Normalizing Flows with stochastic latent variables.
result DIF inherit good computational behavior of NF and can capture distributions with discontinuities.
Continuous-time optimal stopping solved with deep reinforcement learning
problem Optimal stopping problems in continuous time
method CARLOS (Continuous-time Adaptive Reinforcement Learning for Optimal Stopping)
result Higher prices than existing Bermudan solvers, approaching American upper bound
Our goal is to identify the type and number of static equilibrium points of solids arising from fine, equidistant n-discretrizations of smooth, convex surfaces. We assume uniform gravity and a frictionless, horizontal, planar support. We show that as n approaches infinity these numbers fluctuate around specific val…
Physics-informed neural operator learns from coarse to fine discretized data.
problem Lack of high-fidelity training data and uneven grid resolution.
method Physics-informed multi-resolution neural operator framework.
result Learn from arbitrarily discretized input functions using latent embedding and finite difference solver.
Architecture optimization, which is a technique for finding an efficient neural network that meets certain requirements, generally reduces to a set of multiple-choice selection problems among alternative sub-structures or parameters. The discrete nature of the selection problem, however, makes this optimization difficu…
EBMs trained on discrete data using heat equations on graph structures.
problem Training EBMs on discrete or mixed data.
method Heat equations on graph structures for data perturbation.
result Efficacy demonstrated in various applications.
Parameters in deep neural networks which are trained on large-scale databases can generalize across multiple domains, which is referred as "transferability". Unfortunately, the transferability is usually defined as discrete states and it differs with domains and network architectures. Existing works usually heuristical…
Recent neural text-to-speech (TTS) models with fine-grained latent features enable precise control of the prosody of synthesized speech. Such models typically incorporate a fine-grained variational autoencoder (VAE) structure, extracting latent features at each input token (e.g., phonemes). However, generating samples …
We provide a detailed characterization of the optimal consumption stream for the additive habit-forming utility maximization problem, in a framework of general discrete-time incomplete markets and random endowments. This characterization allows us to derive the monotonicity and concavity of the optimal consumption as a…
HiPPO-Prophecy models can learn dynamical systems without fine-tuning.
problem Learning dynamical systems in context without fine-tuning parameters.
method Introduced a novel weight construction for SSMs that approximates derivatives of input signals.
result Discrete SSMs can predict the next state of any dynamical system after observing previous states.
Generative models with both discrete and continuous latent variables are highly motivated by the structure of many real-world data sets. They present, however, subtleties in training often manifesting in the discrete latent being under leveraged. In this paper, we show that such models are more amenable to training whe…
POUF fine-tunes large models without labeled data.
problem Lack of labeled data for fine-tuning large pre-trained models.
method Prompt-oriented unsupervised fine-tuning.
result Consistent improvements across various tasks.
Current state-of-the-art discrete optimization methods struggle behind when it comes to challenging contrast-enhancing discrete energies (i.e., favoring different labels for neighboring variables). This work suggests a multiscale approach for these challenging problems. Deriving an algebraic representation allows us to…
New method optimizes diffusion models without fine-tuning, integrating soft value functions.
problem Optimizing natural design spaces of images, molecules, DNA, RNA, and protein sequences.
method Iterative sampling method integrating soft value functions into diffusion model inference.
result Directly utilizes non-differentiable features/reward feedback, applies to discrete diffusion models.
Study high-frequency trading game with price impact, finding unique equilibrium.
problem Optimal execution in a trading game with transient price impact.
method Analyzes high-frequency limit of an n-trader optimal execution game. result High-frequency limit converges to a continuous-time model with quadratic costs.
Study uses multi-task Bayesian optimization to speed up SVM hyperparameter tuning for nodules diagnosis.
problem Redundant and time-consuming hyperparameter tuning for SVM classifiers in medical imaging.
method Employed multi-task Bayesian optimization to accelerate hyperparameter search.
result Multi-task Bayesian optimization significantly accelerates hyperparameter search.
The paper proposes DEA to make graph neural networks fairer in link prediction.
problem Graph neural networks can unfairly prioritize certain social groups in link prediction.
method Drop Edges and Adapt (DEA) fine-tuning strategy with covariance constraints.
result DEA improves fairness and accuracy in link prediction tasks.
Investigates the relationship between ResNets and Neural ODEs, quantifying their closeness and providing training methods.
problem Quantifying the distance between ResNet dynamics and Neural ODE solutions.
method Bounding the distance between hidden state trajectories and Neural ODE solutions, using gradient descent and Heun's method.
result Gradient descent and Heun's method can implicitly regularize ResNets towards Neural ODEs, especially for smooth residual functions.
The study of topological groups with compact open subgroups and their geometric properties.
problem Characterizing and understanding topological groups with compact open subgroups.
method Geometric techniques, discrete actions on complexes, quasi-isometry invariance, and hyperbolic fine graphs.
result Generalizations of discrete group results to topological groups with compact open subgroups.
Sig-DEG speeds up diffusion models by distilling them into faster approximations.
problem Computational intensity of diffusion models at inference time.
method Signature-based differential equation generation to summarize Brownian motion.
result Sig-DEG reduces inference steps by an order of magnitude while maintaining generation quality.
In recent years, dock-less shared bikes have been widely spread across many cities in China and facilitate people's lives. However, at the same time, it also raises many problems about dock-less shared bike management due to the mismatching between demands and real distribution of bikes. Before deploying dock-less shar…
New method reduces fine-tuning cost for reused models.
problem Repeating fine-tuning costs with outdated foundation models.
method Portable Reward Tuning (PRT) trains a reward model to maximize the same loss function as fine-tuning.
result PRT achieves comparable accuracy to inference-time tuning with less inference cost.
Proposes a differentiable STFT for more efficient optimization of hop length.
problem Efficient optimization of hop length in STFT for better temporal control.
method Introduces a differentiable version of STFT with continuous hop length.
result Improves optimization methods like gradient descent for STFT.
Physics-informed neural networks (PINNs) encode physical conservation laws and prior physical knowledge into the neural networks, ensuring the correct physics is represented accurately while alleviating the need for supervised learning to a great degree. While effective for relatively short-term time integration, when …
Study proposes a multi-agent system using LLMs for REIT trading, outperforming benchmarks.
problem Low-volatility Chinese REIT market, low risk-adjusted returns.
method Multi-agent framework with four types of agents, prediction model pathways, fine-tuning.
result Multi-agent strategies outperform buy-and-hold in terms of return, Sharpe ratio, and drawdown.
Paper finds sparse representation of functions using inverse scale space flow.
problem Finding sparse representation of L2 functions. method Inverse scale space flow to minimize L2 loss. result Convergence to optimal solution in ideal and noisy cases.
We present a local formulation for 2D Discrete Exterior Calculus (DEC) similar to that of the Finite Element Method (FEM), which allows a natural treatment of material heterogeneity (element by element). It also allows us to deduce, in a robust manner, anisotropic fluxes and the DEC discretization of the pullback of 1-…
FGSM is more stable in adversarially robust transfer learning than PGD.
problem Computational efficiency in adversarially robust transfer learning.
method Revisited use of FGSM in adversarial fine-tuning.
result FGSM is more stable and efficient in adversarial fine-tuning.
The paper connects discrete choice models to multi-armed bandit algorithms with sublinear regret bounds.
problem Optimizing user choices in a multi-armed bandit setting.
method Establishes connections between discrete choice models and multi-armed bandit algorithms, providing sublinear regret bounds and novel algorithms.
result Sublinear regret bounds for a family of algorithms, including the Exp3 algorithm.
Improved forecast accuracy for energy systems through decision-focused fine-tuning.
problem Challenges in integrating forecast values into time series models for diverse and specific instances.
method Decision-focused fine-tuning within time series foundation models for dispatchable feeder optimization.
result Improvement of 9.45% in average total daily costs.
LoRA-Curve connects independent LoRA optima through continuous low-loss valleys, improving Bayesian model averaging.
problem Challenges in estimating epistemic uncertainty in LoRA-based Bayesian inference.
method Introduces LoRA-Curve, a segmented Bézier curve parameterization in the LoRA space, with free and anchored configurations.
result Empirically shows that connecting independent LoRA optima through continuous low-loss valleys improves mutual information of the predictive distribution.
AdaCat improves density estimation and planning in autoregressive models.
problem Efficiently modeling sharp density changes in continuous data.
method Adaptive Categorical Discretization (AdaCat) for autoregressive models.
result Improves density estimation and planning in various data types.
Space mapping speeds up shape optimization for PDEs.
problem Efficiently solving shape optimization problems constrained by PDEs.
method Combines fine and coarse model optimizations using Riemannian metrics.
result Space mapping methods are highly efficient for complex shape optimization problems.
A new framework for controllable generation of discrete masked models.
problem Efficient controllable generation of discrete data models.
method Plug-and-play framework based on importance sampling.
result Demonstrates versatility across multiple domains, including protein design.
Fine-tuning a time series model improves financial price prediction accuracy.
problem Improving accuracy in predicting financial market prices using large models.
method Continual pre-training of a time series foundation model on financial data to fine-tune its performance for price prediction.
result The fine-tuned model outperforms the baseline in various financial metrics.
A multilevel optimization method for constrained problems.
problem Regularized constrained linear inverse problems with box constraints.
method Geometric multilevel optimization with varying discretization levels.
result Preserves feasibility of updates while speeding up computations.
MixFT re-partitions data into sub-domains for better TSFM fine-tuning.
problem Improving zero-shot forecasting for new time series domains.
method MixFT re-divides data using Bayesian mixtures into homogeneous sub-domains for separate fine-tuning.
result MixFT outperforms per-dataset fine-tuning methods.
SHIFT method optimally estimates heterogeneous discrete distributions with limited communication.
problem Collaborative learning of discrete distributions under heterogeneity and communication constraints.
method Two-stage method: First, users learn a central distribution; then, fine-tune this to estimate individual distributions.
result SHIFT is minimax optimal in the model of heterogeneity and under communication constraints.
Generative profiling improves real-time task timing for varied resource contexts.
problem Inaccurate task timing analysis for complex hardware architectures.
method Nonparametric, conditional multi-marginal Schrödinger Bridge (MSB) formulation for synthesizing context-dependent timing profiles.
result Maximum likelihood accurate execution profiles for unseen resource contexts.
This study introduces balanced DRPS and OrderedLogitNN for better QDE of discrete-level questions.
problem Lack of ordinal regression methods and fair evaluation metrics for discrete-level QDE.
method Introduces balanced DRPS and OrderedLogitNN, fine-tunes BERT on RACE++ and ARC datasets.
result OrderedLogitNN outperforms other models on complex QDE tasks.