Machine Learning

SB-GFlowNets: Streaming Bayesian inference is now efficient and accurate using GFlowNets, enabling real-time model updates for large, sequential datasets.

Incentive-aware algorithm achieves low regret in strategic multi-armed bandits under debt-free reporting, establishing truthful equilibrium among arms.

This paper presents a novel probabilistic regret bound (PRB) framework for Bayesian optimization, replacing the traditional fixed-budget stopping rule with a criterion based on the probability of find…

Bias-free performative prediction is achieved using a novel lazy deployment scheme with SGD, handling non-convex loss functions.

Researchers extended stochastic optimal control theory to infinite-dimensional spaces, enabling the creation of diffusion bridges for generative modeling in function spaces, demonstrating applications…

Deep kernel machines now achieve 94.5% accuracy on CIFAR-10, matching neural networks, by using stochastic kernel regularization to improve generalization.

Contextual bandits with graph feedback achieve near-optimal regret by leveraging a novel graph-theoretic quantity that interpolates between independence and maximum acyclic subgraph numbers, depending…

Adaptive learning rate tuners often underperform; Curvature Dynamics Aware Tuning (CDAT) prioritizes long-term curvature stabilization, outperforming tuned constant learning rates.

On-device training with fixed-point forward gradients enables efficient model personalization on resource-constrained edge devices, overcoming backpropagation’s memory limitations.

State-free Reinforcement Learning (SFRL) framework eliminates the need for state-space information in RL algorithms, achieving regret bounds independent of the state space size and adaptive to the rea…

GRAPHSSM: a novel graph state space model efficiently captures temporal graph dynamics, overcoming limitations of existing sequence models.

State Chrono Representation (SCR) enhances reinforcement learning generalization by incorporating extensive temporal information and cumulative rewards into state representations, improving performanc…

STk: a novel, differentiable module solves Top-k problems in neural networks without extra time/GPU memory, boosting performance in long-tailed learning.

SPRINQL: Sub-optimal Demonstrations for Offline Imitation Learning

SPO: A novel model-based RL algorithm leverages parallelisable Monte Carlo tree search for efficient and robust policy improvement in both discrete and continuous environments.

STMixer: a novel SNN architecture enabling high performance on both synchronous and asynchronous neuromorphic hardware, achieving comparable results to spiking transformers with drastically lower powe…

Spiking Graph Neural Networks (SGNNs) on Riemannian Manifolds achieve superior performance and energy efficiency via a novel Manifold Spiking GNN (MSG).

Speculative MCTS accelerates AlphaZero training by implementing speculative execution, enabling parallel processing of future moves and reducing latency by up to 5.8x.

SRCPO: a novel spectral risk measure-constrained RL algorithm guaranteeing convergence to a global optimum, outperforming existing methods in continuous control tasks.

PGLDM, a novel algorithm, accurately identifies shared and private dynamics in two generalized-linear time series, improving model accuracy and enabling lower-dimensional latent state representations.