相关论文: Predictive Modeling of Rat Brain Local Field Poten…

We present a reward-predictive, model-based deep learning method featuring trajectory-constrained visual attention for local planning in visual navigation tasks. Our method learns to place visual attention at locations in latent image space…

Deep feed-forward convolutional neural networks (CNNs) have become ubiquitous in virtually all machine learning and computer vision challenges; however, advancements in CNNs have arguably reached an engineering saturation point where…

Multi-electrode arrays covering several square millimeters of neural tissue provide simultaneous access to population signals such as extracellular potentials and spiking activity of one hundred or more individual neurons. The…

Designing reward functions for continuous-control robotics often leads to subtle misalignments or reward hacking, especially in complex tasks. Preference-based RL mitigates some of these pitfalls by learning rewards from comparative…

Robotic real-world reinforcement learning (RL) with vision-language-action (VLA) models is bottlenecked by sparse, handcrafted rewards and inefficient exploration. We introduce VLAC, a general process reward model built upon InternVL and…

Reward design is central to reinforcement learning from human feedback (RLHF) and alignment research. In this work, we propose a unified framework to study hard, continuous, and hybrid reward structures for fine-tuning large language models…

Reinforcement Learning from Human Feedback (RLHF) reward models exhibit systematic failures on longtail distributions, leading to reward hacking and misalignment. We propose a mechanistic interpretability framework that identifies…

Every RLHF-trained language model is shaped by a reward model, yet the mechanistic interpretability toolkit -- logit lens, direct logit attribution, activation patching, sparse autoencoders -- was built for generative LLMs whose primitives…

An appropriate reward function is of paramount importance in specifying a task in reinforcement learning (RL). Yet, it is known to be extremely challenging in practice to design a correct reward function for even simple tasks.…

Continuous neural field models with inhomogeneous synaptic connectivities are known to support traveling fronts as well as stable bumps of localized activity. We analyze stationary localized structures in a neural field model with periodic…

Feedback-rich neural architectures can regenerate earlier representations and inject temporal context, making them a natural setting for strictly local synaptic plasticity. Existing literature raises doubt about whether a minimal,…

Reinforcement learning (RL) algorithms allow artificial agents to improve their selection of actions to increase rewarding experiences in their environments. Temporal Difference (TD) Learning -- a model-free RL method -- is a leading…

Large Vision-Language Models (LVLMs) show promise for embodied planning tasks but struggle with complex scenarios involving unfamiliar environments and multi-step goals. Current approaches rely on environment-agnostic imitation learning…

Neural field models represent neuronal communication on a population level via synaptic weight functions. Using voltage sensitive dye (VSD) imaging it is possible to obtain measurements of neural fields with a relatively high spatial and…

We propose Wake-Sleep Consolidated Learning (WSCL), a learning strategy leveraging Complementary Learning System theory and the wake-sleep phases of the human brain to improve the performance of deep neural networks for visual…

Machine Learning (ML) interatomic models and potentials have been widely employed in simulations of materials. Long-range interactions often dominate in some ionic systems whose dynamics behavior is significantly influenced. However, the…

General-purpose trajectory planning algorithms for automated driving utilize complex reward functions to perform a combined optimization of strategic, behavioral, and kinematic features. The specification and tuning of a single reward…

A reliable reward model is essential for aligning large language models with human preferences through reinforcement learning from human feedback. However, standard reward models are susceptible to spurious features that are not causally…

The aim of this paper is to study the reward based policy exploration problem in a supervised learning approach and enable robots to form complex movement trajectories in challenging reward settings and search spaces. For this, the…

This paper presents a new artificial neuron model capable of learning its receptive field in the topological domain of inputs. The model provides adaptive and differentiable local connectivity (plasticity) applicable to any domain. It…