New algorithm achieves both static and dynamic regret optimally against an oblivious adversary for deterministic losses.
problem Achieving optimal static and dynamic regret simultaneously in adversarial bandits.
method Extends impossibility result to deterministic losses, uses negative static regret and Blackwell approachability.
result First algorithm achieving optimal static and dynamic regret simultaneously against an oblivious adversary.
This work finds mixed equilibria in machine learning problems using measures and simultaneous gradient ascent-descent.
problem Finding pure equilibria in machine learning problems is computationally hard.
method Entropic regularization, simultaneous gradient ascent-descent, and particle discretization in the Wasserstein metric.
result Global convergence towards the global equilibrium in mixed equilibria problems.
Paper proposes algorithms to minimize both dynamic and adaptive regret simultaneously.
problem Traditional regret minimization algorithms are suboptimal for changing environments.
method Developed novel online algorithms to minimize dynamic and adaptive regret simultaneously.
result Proposed algorithms minimize dynamic and adaptive regret over any interval.
Neuroscience is experiencing a data revolution in which many hundreds or thousands of neurons are recorded simultaneously. Currently, there is little consensus on how such data should be analyzed. Here we introduce LFADS (Latent Factor Analysis via Dynamical Systems), a method to infer latent dynamics from simultaneous…
Study forecasts stock returns on JSE using SGDLMs capturing cross-series dependencies.
problem Accurate forecasting of multivariate time series data.
method Simultaneous Graphical Dynamic Linear Models (SGDLMs) with customised DLMs and importance sampling/mean-field variational Bayes.
result SGDLMs accurately forecast stock data on JSE and respond to market changes.
The paper develops methods to estimate optimal treatment sequences under policy constraints.
problem Estimating the best sequence of treatments over multiple stages for individuals.
method Empirical welfare maximization approach, solving treatment assignment sequentially or simultaneously.
result Established convergence rates and upper bounds for estimation methods.
Paper analyzes convergence of dynamic policy gradient for MDPs, improving performance in finite-time problems.
problem Optimal policies in finite-time MDPs are not stationary and require epoch-specific training.
method Introduces dynamic policy gradient combining dynamic programming and policy gradient, analyzes convergence for softmax parametrisation.
result Dynamic policy gradient training exploits finite-time structure, leading to better convergence bounds.
Deep learning solves and estimates complex financial models.
problem Estimating and solving continuous-time financial models.
method Uses deep learning to solve and estimate models simultaneously.
result Demonstrates advantages like generality and large state space handling.
Inverse Reinforcement Learning (IRL) describes the problem of learning an unknown reward function of a Markov Decision Process (MDP) from observed behavior of an agent. Since the agent's behavior originates in its policy and MDP policies depend on both the stochastic system dynamics as well as the reward function, the …
Games generalize the single-objective optimization paradigm by introducing different objective functions for different players. Differentiable games often proceed by simultaneous or alternating gradient updates. In machine learning, games are gaining new importance through formulations like generative adversarial netwo…
Neurons in cortical circuits exhibit coordinated spiking activity, and can produce correlated synchronous spikes during behavior and cognition. We recently developed a method for estimating the dynamics of correlated ensemble activity by combining a model of simultaneous neuronal interactions (e.g., a spin-glass model)…
Proposes dynamic channel pruning during neural network training.
problem Pruning neural networks during training to reduce computational cost and improve efficiency.
method Dynamic channel propagation to update channel utility values and selectively prune channels.
result Our scheme trains and prunes neural networks simultaneously, achieving superior performance.
Min-max formulations have attracted great attention in the ML community due to the rise of deep generative models and adversarial methods, while understanding the dynamics of gradient algorithms for solving such formulations has remained a grand challenge. As a first step, we restrict to bilinear zero-sum games and giv…
Study stability of trading strategy under market perturbations.
problem Dynamic stability of trading strategy under market changes.
method Established reverse conjugacy characterizations, proved continuity and convergence of indirect utility process.
result Continuity and first-order convergence of indirect utility process under market perturbations.
SimCD simultaneously clusters cells and identifies differential gene expression in scRNA-seq data.
problem Separate clustering and differential expression analysis for scRNA-seq data leads to suboptimal results.
method Develops SimCD, a unified hierarchical gamma-negative binomial model for simultaneous cell clustering and differential expression analysis.
result SimCD outperforms existing methods in discovering cell clusters and capturing dynamic expression changes.
Optimal switching regret for all segmentations in online convex optimisation.
problem Non-stationary online convex optimisation problems.
method Developed an efficient algorithm to achieve optimal switching regret on every possible segmentation.
result Achieved asymptotically optimal switching regret on every possible segmentation simultaneously.
New RL framework simulates financial market dynamics.
problem Complex financial market dynamics under various scenarios.
method Two RL families learn simultaneously, using Deep RL and parametrized reward.
result Agents learn a shared policy for diverse behaviors.
We make posterior sampling in FWI feasible for large surveys.
problem Uncertainty-aware subsurface models at field scale.
method Coupling diffusion-based posterior sampling with simultaneous-source FWI data.
result Lower model error and better data fit at reduced computational cost.
Learning network representations is a fundamental task for many graph applications such as link prediction, node classification, graph clustering, and graph visualization. Many real-world networks are interpreted as dynamic networks and evolve over time. Most existing graph embedding algorithms were developed for stati…
Model approximates market prices and returns without prior market dynamics.
problem Simultaneously approximate market prices and log returns.
method GDN model of Kratsios and Papon (2022) for generalized Ornstein-Uhlenbeck process.
result Universal approximation guarantees for conditional distributions and contingent claims.
BM2 learns Schrödinger bridges using neural networks.
problem Learning dynamic transport maps between two distributions.
method Coupled Bridge Matching (BM2) with neural networks. result Preliminary theoretical analysis and numerical experiments show BM2's effectiveness. Nonlinear state-space models are powerful tools to describe dynamical structures in complex time series. In a streaming setting where data are processed one sample at a time, simultaneous inference of the state and its nonlinear dynamics has posed significant challenges in practice. We develop a novel online learning f…
Model learns Lagrangian dynamics from images for better prediction and control.
problem Lack of interpretability and applicability to high-dimensional data like images.
method Unsupervised neural network model that learns Lagrangian dynamics from images using a coordinate-aware VAE.
result Model infers interpretable Lagrangian dynamics, enabling long-term prediction and synthesis of controllers.
High dimensional time series are endemic in applications of machine learning such as robotics (sensor data), computational biology (gene expression data), vision (video sequences) and graphics (motion capture data). Practical nonlinear probabilistic approaches to this data are required. In this paper we introduce the v…
In a complex system, the interactions between individual agents often lead to emergent collective behavior like spontaneous synchronization, swarming, and pattern formation. The topology of the network of interactions can have a dramatic influence over those dynamics. In many studies, researchers start with a specific …
Generative adversarial networks (GANs) are notoriously difficult to train and the reasons underlying their (non-)convergence behaviors are still not completely understood. By first considering a simple yet representative GAN example, we mathematically analyze its local convergence behavior in a non-asymptotic way. Furt…
Understanding the learning dynamics of neural networks is one of the key issues for the improvement of optimization algorithms as well as for the theoretical comprehension of why deep neural nets work so well today. In this paper, we introduce a random matrix-based framework to analyze the learning dynamics of a single…
Gaussian processes for dynamical systems with Koopman equivariance.
problem Forecasting and learning representations of nonlinear dynamical systems.
method Koopman-equivariant Gaussian processes with linear time-invariant responses and trajectory-based equivariance.
result Enhanced forecasting performance compared to kernel-based methods.
Instabilities in the price dynamics of a large number of financial assets are a clear sign of systemic events. By investigating a set of 20 high cap stocks traded at the Italian Stock Exchange, we find that there is a large number of high frequency cojumps. We show that the dynamics of these jumps is described neither …
In this paper we consider self-supervised representation learning to improve sample efficiency in reinforcement learning (RL). We propose a forward prediction objective for simultaneously learning embeddings of states and action sequences. These embeddings capture the structure of the environment's dynamics, enabling e…
Many real-world systems studied are governed by complex, nonlinear dynamics. By modeling these dynamics, we can gain insight into how these systems work, make predictions about how they will behave, and develop strategies for controlling them. While there are many methods for modeling nonlinear dynamical systems, exist…
ELS framework improves safety alignment by dynamically steering LLMs towards helpful responses.
problem Over-Refusal in Aligned Large Language Models
method Fine-tuning free framework using an Energy-Based Model (EBM) to dynamically steer LLMs during inference.
result Extensive experiments show a significant reduction in false refusals (from 57.3% to 82.6%) while maintaining safety performance.
DVE models dynamic changes in feature embeddings for better sequence-aware applications.
problem Dynamic characterization of feature variations in time-varying data.
method Dynamic Variational Embedding (DVE) using recurrent neural networks.
result DVE models intrinsic nature and temporal variation of nodes effectively.
Simultaneously recorded electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) can be used to non-invasively measure the spatiotemporal dynamics of the human brain. One challenge is dealing with the artifacts that each modality introduces into the other when the two are recorded concurrently, for…
A method of simultaneously optimizing both the structure of neural networks and the connection weights in a single training loop can reduce the enormous computational cost of neural architecture search. We focus on the probabilistic model-based dynamic neural network structure optimization that considers the probabilit…
In evolving complex systems such as air traffic and social organizations, collective effects emerge from their many components' dynamic interactions. While the dynamic interactions can be represented by temporal networks with nodes and links that change over time, they remain highly complex. It is therefore often neces…
Interacting systems are prevalent in nature, from dynamical systems in physics to complex societal dynamics. The interplay of components can give rise to complex behavior, which can often be explained using a simple model of the system's constituent parts. In this work, we introduce the neural relational inference (NRI…
Modeling bank leverage dynamics to understand systemic risk in financial markets.
problem Understanding systemic risk in financial markets triggered by bank leverage dynamics.
method Developed a dynamical model of bank leverage, analyzing coupled dynamics in isolated and interconnected bank models.
result Identified a procyclical feedback loop between asset prices and leverage, leading to chaotic dynamics.
The health state assessment and remaining useful life (RUL) estimation play very important roles in prognostics and health management (PHM), owing to their abilities to reduce the maintenance and improve the safety of machines or equipment. However, they generally suffer from this problem of lacking prior knowledge to …
A blockchain protocol uses bandit algorithms to dynamically price transactions.
problem Maximizing revenue from decentralized blockchain Indexers competing for queries.
method Dynamic pricing using Gaussian bandits for multiple agents.
result Improved revenue through dynamic pricing in a decentralized blockchain environment.
Optimal online linear regression in dynamic environments using discounted Vovk-Azoury-Warmuth forecaster.
problem Achieving optimal performance in dynamic online linear regression without prior knowledge.
method Developed a discounted variant of the Vovk-Azoury-Warmuth forecaster to achieve optimal dynamic regret guarantees.
result Achieved dynamic regret of the form $O\left(d\log(T)\vee \sqrt{dP_{T}^γ(\vec{u})T}
ight)$, with a learnable discount factor.
The paper simplifies multi-agent RL dynamics in finite-state Markov games using homogenization.
problem Approximating complex multi-agent reinforcement learning dynamics in finite-state Markov games.
method Rescaling learning process by reducing learning rate and increasing update frequency, proving convergence to an ODE.
result The rescaled process converges to an ODE that approximates the agent's learning dynamics.
Universal online optimization for dynamic environments using uniclass prediction.
problem Online optimization in changing environments with dynamic regret.
method Reduces dynamic online optimization to uniclass prediction problem, allowing control over dynamic regret bounds.
result First paper with state-of-the-art dynamic regret guarantees for general convex cost functions.
Proposes a method to learn system dynamics and region of attraction from trajectories.
problem Learning accurate dynamics and region of attraction from system trajectories.
method Uses local stability information as a prior to learn vector field and region of attraction.
result Efficient sampling and accurate estimate of dynamics in inner approximation of region of attraction.
Develops a framework for robust RL with dynamic risk measures.
problem Optimal RL strategies depend on risk preferences and model dynamics.
method Dynamic robust distortion risk measures, Wasserstein ball, neural networks, strictly consistent scoring functions, policy gradient formulae, actor-critic algorithm.
result Demonstrates improved performance in portfolio allocation example.
Study dynamics of automorphisms on cubic surfaces and their connection to Painlevé 6.
problem Dynamics of holomorphic automorphisms on cubic surfaces and their relation to Painlevé 6.
method Defined Julia and Fatou sets, studied locally discrete and non-discrete dynamics, and proved existence of non-empty Fatou and Julia sets.
result Existence of non-empty Fatou and Julia sets for the group action.
DBGDGM models dynamic brain graphs for better understanding brain function.
problem Previous brain graph models ignore temporal dynamics, limiting their usefulness.
method DBGDGM clusters brain regions into evolving communities and learns dynamic node embeddings.
result DBGDGM outperforms baselines in graph generation, dynamic link prediction, and graph classification.
AR model forecasts partially observed dynamical time series by estimating evolution function and imputing missing variables.
problem Forecasting dynamical time series with missing variables.
method Autoregressive with slack time series (ARS) model.
result ARS model forecasts future time series with time-invariant and linear assumptions.