Bayesian Design Principles for Offline-to-Online Reinforcement Learning

Offline reinforcement learning (RL) is crucial for real-world applications where exploration can be costly or unsafe. However, offline learned policies are often suboptimal, and further online fine-tuning is required. In this paper, we tackle the fundamental dilemma of offline-to-online fine-tuning: if the agent remains pessimistic, it may fail to learn a better policy, while if it becomes optimistic directly, performance may suffer from a sudden drop. We show that Bayesian design principles are crucial in solving such a dilemma. Instead of adopting optimistic or pessimistic policies, the agent should act in a way that matches its belief in optimal policies. Such a probability-matching agent can avoid a sudden performance drop while still being guaranteed to find the optimal policy. Based on our theoretical findings, we introduce a novel algorithm that outperforms existing methods on various benchmarks, demonstrating the efficacy of our approach. Overall, the proposed approach provides a new perspective on offline-to-online RL that has the potential to enable more effective learning from offline data.

Digital-analog hybrid matrix multiplication processor for optical neural networks

The computational demands of modern AI have spurred interest in optical neural networks (ONNs) which offer the potential benefits of increased speed and lower power consumption. However, current ONNs face various challenges,most significantly a limited calculation precision (typically around 4 bits) and the requirement for high-resolution signal format converters (digital-to-analogue conversions (DACs) and analogue-to-digital conversions (ADCs)). These challenges are inherent to their analog computing nature and pose significant obstacles in practical implementation. Here, we propose a digital-analog hybrid optical computing architecture for ONNs, which utilizes digital optical inputs in the form of binary words. By introducing the logic levels and decisions based on thresholding, the calculation precision can be significantly enhanced. The DACs for input data can be removed and the resolution of the ADCs can be greatly reduced. This can increase the operating speed at a high calculation precision and facilitate the compatibility with microelectronics. To validate our approach, we have fabricated a proof-of-concept photonic chip and built up a hybrid optical processor (HOP) system for neural network applications. We have demonstrated an unprecedented 16-bit calculation precision for high-definition image processing, with a pixel error rate (PER) as low as $1.8\times10^{-3}$ at an signal-to-noise ratio (SNR) of 18.2 dB. We have also implemented a convolutional neural network for handwritten digit recognition that shows the same accuracy as the one achieved by a desktop computer. The concept of the digital-analog hybrid optical computing architecture offers a methodology that could potentially be applied to various ONN implementations and may intrigue new research into efficient and accurate domain-specific optical computing architectures for neural networks.

Integration and Performance Analysis of Artificial Intelligence and Computer Vision Based on Deep Learning Algorithms

This paper focuses on the analysis of the application effectiveness of the integration of deep learning and computer vision technologies. Deep learning achieves a historic breakthrough by constructing hierarchical neural networks, enabling end-to-end feature learning and semantic understanding of images. The successful experiences in the field of computer vision provide strong support for training deep learning algorithms. The tight integration of these two fields has given rise to a new generation of advanced computer vision systems, significantly surpassing traditional methods in tasks such as machine vision image classification and object detection. In this paper, typical image classification cases are combined to analyze the superior performance of deep neural network models while also pointing out their limitations in generalization and interpretability, proposing directions for future improvements. Overall, the efficient integration and development trend of deep learning with massive visual data will continue to drive technological breakthroughs and application expansion in the field of computer vision, making it possible to build truly intelligent machine vision systems. This deepening fusion paradigm will powerfully promote unprecedented tasks and functions in computer vision, providing stronger development momentum for related disciplines and industries.

VectorTalker: SVG Talking Face Generation with Progressive Vectorisation

High-fidelity and efficient audio-driven talking head generation has been a key research topic in computer graphics and computer vision. In this work, we study vector image based audio-driven talking head generation. Compared with directly animating the raster image that most widely used in existing works, vector image enjoys its excellent scalability being used for many applications. There are two main challenges for vector image based talking head generation: the high-quality vector image reconstruction w.r.t. the source portrait image and the vivid animation w.r.t. the audio signal. To address these, we propose a novel scalable vector graphic reconstruction and animation method, dubbed VectorTalker. Specifically, for the highfidelity reconstruction, VectorTalker hierarchically reconstructs the vector image in a coarse-to-fine manner. For the vivid audio-driven facial animation, we propose to use facial landmarks as intermediate motion representation and propose an efficient landmark-driven vector image deformation module. Our approach can handle various styles of portrait images within a unified framework, including Japanese manga, cartoon, and photorealistic images. We conduct extensive quantitative and qualitative evaluations and the experimental results demonstrate the superiority of VectorTalker in both vector graphic reconstruction and audio-driven animation.

Joint User Association, Interference Cancellation and Power Control for Multi-IRS Assisted UAV Communications

Intelligent reflecting surface (IRS)-assisted unmanned aerial vehicle (UAV) communications are expected to alleviate the load of ground base stations in a cost-effective way. Existing studies mainly focus on the deployment and resource allocation of a single IRS instead of multiple IRSs, whereas it is extremely challenging for joint multi-IRS multi-user association in UAV communications with constrained reflecting resources and dynamic scenarios. To address the aforementioned challenges, we propose a new optimization algorithm for joint IRS-user association, trajectory optimization of UAVs, successive interference cancellation (SIC) decoding order scheduling and power allocation to maximize system energy efficiency. We first propose an inverse soft-Q learning-based algorithm to optimize multi-IRS multi-user association. Then, SCA and Dinkelbach-based algorithm are leveraged to optimize UAV trajectory followed by the optimization of SIC decoding order scheduling and power allocation. Finally, theoretical analysis and performance results show significant advantages of the designed algorithm in convergence rate and energy efficiency.

C-Silicon-based metasurfaces for aperture-robust spectrometer/imaging with angle integration

Compared with conventional grating-based spectrometers, reconstructive spectrometers based on spectrally engineered filtering have the advantage of miniaturization because of the less demand for dispersive optics and free propagation space. However, available reconstructive spectrometers fail to balance the performance on operational bandwidth, spectral diversity and angular stability. In this work, we proposed a compact silicon metasurfaces based spectrometer/camera. After angle integration, the spectral response of the system is robust to angle/aperture within a wide working bandwidth from 400nm to 800nm. It is experimentally demonstrated that the proposed method could maintain the spectral consistency from F/1.8 to F/4 (The corresponding angle of incident light ranges from 7{\deg} to 16{\deg}) and the incident hyperspectral signal could be accurately reconstructed with a fidelity exceeding 99%. Additionally, a spectral imaging system with 400x400 pixels is also established in this work. The accurate reconstructed hyperspectral image indicates that the proposed aperture-robust spectrometer has the potential to be extended as a high-resolution broadband hyperspectral camera.

Unsupervised Behavior Extraction via Random Intent Priors

Reward-free data is abundant and contains rich prior knowledge of human behaviors, but it is not well exploited by offline reinforcement learning (RL) algorithms. In this paper, we propose UBER, an unsupervised approach to extract useful behaviors from offline reward-free datasets via diversified rewards. UBER assigns different pseudo-rewards sampled from a given prior distribution to different agents to extract a diverse set of behaviors, and reuse them as candidate policies to facilitate the learning of new tasks. Perhaps surprisingly, we show that rewards generated from random neural networks are sufficient to extract diverse and useful behaviors, some even close to expert ones. We provide both empirical and theoretical evidence to justify the use of random priors for the reward function. Experiments on multiple benchmarks showcase UBER's ability to learn effective and diverse behavior sets that enhance sample efficiency for online RL, outperforming existing baselines. By reducing reliance on human supervision, UBER broadens the applicability of RL to real-world scenarios with abundant reward-free data.

Reason for Future, Act for Now: A Principled Framework for Autonomous LLM Agents with Provable Sample Efficiency

Large language models (LLMs) demonstrate impressive reasoning abilities, but translating reasoning into actions in the real world remains challenging. In particular, it remains unclear how to complete a given task provably within a minimum number of interactions with the external environment, e.g., through an internal mechanism of reasoning. To this end, we propose a principled framework with provable regret guarantees to orchestrate reasoning and acting, which we call "reason for future, act for now" (\texttt{RAFA}). Specifically, we design a prompt template for reasoning that learns from the memory buffer and plans a future trajectory over a long horizon ("reason for future"). At each step, the LLM agent takes the initial action of the planned trajectory ("act for now"), stores the collected feedback in the memory buffer, and reinvokes the reasoning routine to replan the future trajectory from the new state. The key idea is to cast reasoning in LLMs as learning and planning in Bayesian adaptive Markov decision processes (MDPs). Correspondingly, we prompt LLMs to form an updated posterior of the unknown environment from the memory buffer (learning) and generate an optimal trajectory for multiple future steps that maximizes a value function (planning). The learning and planning subroutines are performed in an "in-context" manner to emulate the actor-critic update for MDPs. Our theoretical analysis proves that the novel combination of long-term reasoning and short-term acting achieves a $\sqrt{T}$ regret. In particular, the regret bound highlights an intriguing interplay between the prior knowledge obtained through pretraining and the uncertainty reduction achieved by reasoning and acting. Our empirical validation shows that it outperforms various existing frameworks and achieves nearly perfect scores on a few benchmarks.

Active Implicit Reconstruction Using One-Shot View Planning

Active object reconstruction using autonomous robots is gaining great interest. A primary goal in this task is to maximize the information of the object to be reconstructed, given limited on-board resources. Previous view planning methods exhibit inefficiency since they rely on an iterative paradigm based on explicit representations, consisting of (1) planning a path to the next-best view only; and (2) requiring a considerable number of less-gain views in terms of surface coverage. To address these limitations, we integrated implicit representations into the One-Shot View Planning (OSVP). The key idea behind our approach is to use implicit representations to obtain the small missing surface areas instead of observing them with extra views. Therefore, we design a deep neural network, named OSVP, to directly predict a set of views given a dense point cloud refined from an initial sparse observation. To train our OSVP network, we generate supervision labels using dense point clouds refined by implicit representations and set covering optimization problems. Simulated experiments show that our method achieves sufficient reconstruction quality, outperforming several baselines under limited view and movement budgets. We further demonstrate the applicability of our approach in a real-world object reconstruction scenario.

How Many Views Are Needed to Reconstruct an Unknown Object Using NeRF?

Neural Radiance Fields (NeRFs) are gaining significant interest for online active object reconstruction due to their exceptional memory efficiency and requirement for only posed RGB inputs. Previous NeRF-based view planning methods exhibit computational inefficiency since they rely on an iterative paradigm, consisting of (1) retraining the NeRF when new images arrive; and (2) planning a path to the next best view only. To address these limitations, we propose a non-iterative pipeline based on the Prediction of the Required number of Views (PRV). The key idea behind our approach is that the required number of views to reconstruct an object depends on its complexity. Therefore, we design a deep neural network, named PRVNet, to predict the required number of views, allowing us to tailor the data acquisition based on the object complexity and plan a globally shortest path. To train our PRVNet, we generate supervision labels using the ShapeNet dataset. Simulated experiments show that our PRV-based view planning method outperforms baselines, achieving good reconstruction quality while significantly reducing movement cost and planning time. We further justify the generalization ability of our approach in a real-world experiment.