亚洲男人的天堂2018av,欧美草比,久久久久久免费视频精选,国色天香在线看免费,久久久久亚洲av成人片仓井空

Improving the multi-step reasoning ability of large language models (LLMs) with offline reinforcement learning (RL) is essential for quickly adapting them to complex tasks. While Direct Preference Optimization (DPO) has shown promise in aligning LLMs with human preferences, it is less suitable for multi-step reasoning tasks because (1) DPO relies on paired preference data, which is not readily available for multi-step reasoning tasks, and (2) it treats all tokens uniformly, making it ineffective for credit assignment in multi-step reasoning tasks, which often come with sparse reward. In this work, we propose OREO (Offline Reasoning Optimization), an offline RL method for enhancing LLM multi-step reasoning. Building on insights from previous works of maximum entropy reinforcement learning, it jointly learns a policy model and value function by optimizing the soft Bellman Equation. We show in principle that it reduces the need to collect pairwise data and enables better credit assignment. Empirically, OREO surpasses existing offline learning methods on multi-step reasoning benchmarks, including mathematical reasoning tasks (GSM8K, MATH) and embodied agent control (ALFWorld). The approach can be extended to a multi-iteration framework when additional resources are available. Furthermore, the learned value function can be leveraged to guide the tree search for free, which can further boost performance during test time.

This study explores the application of the rate-splitting multiple access (RSMA) technique, vital for interference mitigation in modern communication systems. It investigates the use of precoding methods in RSMA, especially in complex multiple-antenna interference channels, employing deep reinforcement learning. The aim is to optimize precoders and power allocation for common and private data streams involving multiple decision-makers. A multi-agent deep deterministic policy gradient (MADDPG) framework is employed to address this complexity, where decentralized agents collectively learn to optimize actions in a continuous policy space. We also explore the challenges posed by imperfect channel side information at the transmitter. Additionally, decoding order estimation is addressed to determine the optimal decoding sequence for common and private data sequences. Simulation results demonstrate the effectiveness of the proposed RSMA method based on MADDPG, achieving the upper bound in single-antenna scenarios and closely approaching theoretical limits in multi-antenna scenarios. Comparative analysis shows superiority over other techniques such as MADDPG without rate-splitting, maximal ratio transmission (MRT), zero-forcing (ZF), and leakage-based precoding methods. These findings highlight the potential of deep reinforcement learning-driven RSMA in reducing interference and enhancing system performance in communication systems.

Large Language Models (LLMs) possess vast amounts of knowledge within their parameters, prompting research into methods for locating and editing this knowledge. Previous work has largely focused on locating entity-related (often single-token) facts in smaller models. However, several key questions remain unanswered: (1) How can we effectively locate query-relevant neurons in decoder-only LLMs, such as Llama and Mistral? (2) How can we address the challenge of long-form (or free-form) text generation? (3) Are there localized knowledge regions in LLMs? In this study, we introduce Query-Relevant Neuron Cluster Attribution (QRNCA), a novel architecture-agnostic framework capable of identifying query-relevant neurons in LLMs. QRNCA allows for the examination of long-form answers beyond triplet facts by employing the proxy task of multi-choice question answering. To evaluate the effectiveness of our detected neurons, we build two multi-choice QA datasets spanning diverse domains and languages. Empirical evaluations demonstrate that our method outperforms baseline methods significantly. Further, analysis of neuron distributions reveals the presence of visible localized regions, particularly within different domains. Finally, we show potential applications of our detected neurons in knowledge editing and neuron-based prediction.

As artificial intelligence (AI) continues advancing, ensuring positive societal impacts becomes critical, especially as AI systems become increasingly ubiquitous in various aspects of life. However, developing "AI for good" poses substantial challenges around aligning systems with complex human values. Presently, we lack mature methods for addressing these challenges. This article presents and evaluates the Positive AI design method aimed at addressing this gap. The method provides a human-centered process to translate wellbeing aspirations into concrete practices. First, we explain the method's four key steps: contextualizing, operationalizing, optimizing, and implementing wellbeing supported by continuous measurement for feedback cycles. We then present a multiple case study where novice designers applied the method, revealing strengths and weaknesses related to efficacy and usability. Next, an expert evaluation study assessed the quality of the resulting concepts, rating them moderately high for feasibility, desirability, and plausibility of achieving intended wellbeing benefits. Together, these studies provide preliminary validation of the method's ability to improve AI design, while surfacing areas needing refinement like developing support for complex steps. Proposed adaptations such as examples and evaluation heuristics could address weaknesses. Further research should examine sustained application over multiple projects. This human-centered approach shows promise for realizing the vision of 'AI for Wellbeing' that does not just avoid harm, but actively benefits humanity.

Adversarial attacks pose significant challenges in 3D object recognition, especially in scenarios involving multi-view analysis where objects can be observed from varying angles. This paper introduces View-Invariant Adversarial Perturbations (VIAP), a novel method for crafting robust adversarial examples that remain effective across multiple viewpoints. Unlike traditional methods, VIAP enables targeted attacks capable of manipulating recognition systems to classify objects as specific, pre-determined labels, all while using a single universal perturbation. Leveraging a dataset of 1,210 images across 121 diverse rendered 3D objects, we demonstrate the effectiveness of VIAP in both targeted and untargeted settings. Our untargeted perturbations successfully generate a singular adversarial noise robust to 3D transformations, while targeted attacks achieve exceptional results, with top-1 accuracies exceeding 95% across various epsilon values. These findings highlight VIAPs potential for real-world applications, such as testing the robustness of 3D recognition systems. The proposed method sets a new benchmark for view-invariant adversarial robustness, advancing the field of adversarial machine learning for 3D object recognition.

Recent contrastive representation learning methods rely on estimating mutual information (MI) between multiple views of an underlying context. E.g., we can derive multiple views of a given image by applying data augmentation, or we can split a sequence into views comprising the past and future of some step in the sequence. Contrastive lower bounds on MI are easy to optimize, but have a strong underestimation bias when estimating large amounts of MI. We propose decomposing the full MI estimation problem into a sum of smaller estimation problems by splitting one of the views into progressively more informed subviews and by applying the chain rule on MI between the decomposed views. This expression contains a sum of unconditional and conditional MI terms, each measuring modest chunks of the total MI, which facilitates approximation via contrastive bounds. To maximize the sum, we formulate a contrastive lower bound on the conditional MI which can be approximated efficiently. We refer to our general approach as Decomposed Estimation of Mutual Information (DEMI). We show that DEMI can capture a larger amount of MI than standard non-decomposed contrastive bounds in a synthetic setting, and learns better representations in a vision domain and for dialogue generation.

Recently, various auxiliary tasks have been proposed to accelerate representation learning and improve sample efficiency in deep reinforcement learning (RL). However, existing auxiliary tasks do not take the characteristics of RL problems into consideration and are unsupervised. By leveraging returns, the most important feedback signals in RL, we propose a novel auxiliary task that forces the learnt representations to discriminate state-action pairs with different returns. Our auxiliary loss is theoretically justified to learn representations that capture the structure of a new form of state-action abstraction, under which state-action pairs with similar return distributions are aggregated together. In low data regime, our algorithm outperforms strong baselines on complex tasks in Atari games and DeepMind Control suite, and achieves even better performance when combined with existing auxiliary tasks.

Benefit from the quick development of deep learning techniques, salient object detection has achieved remarkable progresses recently. However, there still exists following two major challenges that hinder its application in embedded devices, low resolution output and heavy model weight. To this end, this paper presents an accurate yet compact deep network for efficient salient object detection. More specifically, given a coarse saliency prediction in the deepest layer, we first employ residual learning to learn side-output residual features for saliency refinement, which can be achieved with very limited convolutional parameters while keep accuracy. Secondly, we further propose reverse attention to guide such side-output residual learning in a top-down manner. By erasing the current predicted salient regions from side-output features, the network can eventually explore the missing object parts and details which results in high resolution and accuracy. Experiments on six benchmark datasets demonstrate that the proposed approach compares favorably against state-of-the-art methods, and with advantages in terms of simplicity, efficiency (45 FPS) and model size (81 MB).

Graph neural networks (GNNs) are a popular class of machine learning models whose major advantage is their ability to incorporate a sparse and discrete dependency structure between data points. Unfortunately, GNNs can only be used when such a graph-structure is available. In practice, however, real-world graphs are often noisy and incomplete or might not be available at all. With this work, we propose to jointly learn the graph structure and the parameters of graph convolutional networks (GCNs) by approximately solving a bilevel program that learns a discrete probability distribution on the edges of the graph. This allows one to apply GCNs not only in scenarios where the given graph is incomplete or corrupted but also in those where a graph is not available. We conduct a series of experiments that analyze the behavior of the proposed method and demonstrate that it outperforms related methods by a significant margin.

Multi-relation Question Answering is a challenging task, due to the requirement of elaborated analysis on questions and reasoning over multiple fact triples in knowledge base. In this paper, we present a novel model called Interpretable Reasoning Network that employs an interpretable, hop-by-hop reasoning process for question answering. The model dynamically decides which part of an input question should be analyzed at each hop; predicts a relation that corresponds to the current parsed results; utilizes the predicted relation to update the question representation and the state of the reasoning process; and then drives the next-hop reasoning. Experiments show that our model yields state-of-the-art results on two datasets. More interestingly, the model can offer traceable and observable intermediate predictions for reasoning analysis and failure diagnosis.