Reincarnating Reinforcement Learning: Reusing Prior Computation to Accelerate Progress
Representational ethical model calibration
Robert Carruthers
Isabel Straw
James K. Ruffle
Daniel Herron
Amy Nelson
Delmiro Fernandez-Reyes
Geraint Rees
Parashkev Nachev
Revisiting Heterophily For Graph Neural Networks
Sitao Luan
Chenqing Hua
Qincheng Lu
Jiaqi Zhu
Harry Zhao
Mingde Zhao
Shuyuan Zhang
Xiao-Wen Chang
Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using graph structures based on the relational inductive bias (homophily … (voir plus)assumption). While GNNs have been commonly believed to outperform NNs in real-world tasks, recent work has identified a non-trivial set of datasets where their performance compared to NNs is not satisfactory. Heterophily has been considered the main cause of this empirical observation and numerous works have been put forward to address it. In this paper, we first revisit the widely used homophily metrics and point out that their consideration of only graph-label consistency is a shortcoming. Then, we study heterophily from the perspective of post-aggregation node similarity and define new homophily metrics, which are potentially advantageous compared to existing ones. Based on this investigation, we prove that some harmful cases of heterophily can be effectively addressed by local diversification operation. Then, we propose the Adaptive Channel Mixing (ACM), a framework to adaptively exploit aggregation, diversification and identity channels node-wisely to extract richer localized information for diverse node heterophily situations. ACM is more powerful than the commonly used uni-channel framework for node classification tasks on heterophilic graphs and is easy to be implemented in baseline GNN layers. When evaluated on 10 benchmark node classification tasks, ACM-augmented baselines consistently achieve significant performance gain, exceeding state-of-the-art GNNs on most tasks without incurring significant computational burden.
Riemannian Diffusion Models
Chin-Wei Huang
Milad Aghajohari
Diffusion models are recent state-of-the-art methods for image generation and likelihood estimation. In this work, we generalize continuous-… (voir plus)time diffusion models to arbitrary Riemannian manifolds and derive a variational framework for likelihood estimation. Computationally, we propose new methods for computing the Riemannian divergence which is needed for likelihood estimation. Moreover, in generalizing the Euclidean case, we prove that maximizing this variational lower-bound is equivalent to Riemannian score matching. Empirically, we demonstrate the expressive power of Riemannian diffusion models on a wide spectrum of smooth manifolds, such as spheres, tori, hyperboloids, and orthogonal groups. Our proposed method achieves new state-of-the-art likelihoods on all benchmarks.
Robust Policy Learning over Multiple Uncertainty Sets
Annie Xie
Shagun Sodhani
Chelsea Finn
Amy Zhang
Reinforcement learning (RL) agents need to be robust to variations in safety-critical environments. While system identification methods prov… (voir plus)ide a way to infer the variation from online experience, they can fail in settings where fast identification is not possible. Another dominant approach is robust RL which produces a policy that can handle worst-case scenarios, but these methods are generally designed to achieve robustness to a single uncertainty set that must be specified at train time. Towards a more general solution, we formulate the multi-set robustness problem to learn a policy robust to different perturbation sets. We then design an algorithm that enjoys the benefits of both system identification and robust RL: it reduces uncertainty where possible given a few interactions, but can still act robustly with respect to the remaining uncertainty. On a diverse set of control tasks, our approach demonstrates improved worst-case performance on new environments compared to prior methods based on system identification and on robust RL alone.
Robustness of Whittle Index Policy to Model Approximation
Amit Sinha
Scalable Operator Allocation for Multirobot Assistance: A Restless Bandit Approach
Abhinav Dahiya
Nima Akbarzadeh
Stephen L. Smith
In this article, we consider the problem of allocating human operators in a system with multiple semiautonomous robots. Each robot is requir… (voir plus)ed to perform an independent sequence of tasks, subject to a chance of failing and getting stuck in a fault state at every task. If and when required, a human operator can assist or teleoperate a robot. Conventional dynamic programming-based techniques used to solve such problems face scalability issues due to an exponential growth of state and action spaces with the number of robots and operators. In this article, we derive conditions under which the operator allocation problem satisfies a technical condition called indexability, thereby enabling the use of the Whittle index heuristic. The conditions are easy to check, and we show that they hold for a wide range of problems of interest. Our key insight is to leverage the structure of the value function of individual robots, resulting in conditions that can be verified separately for each state of each robot. We apply these conditions to two types of transitions commonly seen in remote robot supervision systems. Through numerical simulations, we demonstrate the efficacy of Whittle index policy as a near-optimal and scalable approach that outperforms existing scalable methods.
Scaling the Number of Tasks in Continual Learning
Timothee LESORT
Oleksiy Ostapenko
Diganta Misra
Md Rifat Arefin
Pau Rodriguez
Sociotechnical Harms: Scoping a Taxonomy for Harm Reduction
Renee Shelby
Shalaleh Rismani
Kathryn Henne
Paul Nicholas
N'mah Fodiatu Yilla
Jess Gallegos
Andrew J Smart
Emilio Garcia
Gurleen Virk
Source-summary Entity Aggregation in Abstractive Summarization.
José-ángel González
Annie Priyadarshini Louis
A Synchro-Set-Aided Breadth-First Sphere Decoder for Polar-Coded MIMO Systems
Huayi Zhou
Xiangyun Deng
Yiqian Cai
Yifei Shen
Minhua Yang
Xiaohu You
Chuan Zhang
The joint optimization of multiple-input-multiple-output (MIMO) detection and polar decoding has become a research hotspot for future commun… (voir plus)ication systems. The error-correction performance of the separate detection and decoding (SDD) is far from the Shannon capacity, which cannot meet the requirements of communication scenarios such as ultra-reliable and low latency communications (URLLC). The existing joint detection and decoding (JDD) using breadth-first sphere decoding (BFSD) improves the reliability over SDD but still has a huge performance loss on low-rate codes. In this paper, JDD using synchro-set-aided BFSD (SA-BFSD) is proposed to greatly improve the error-correction performance for polar-coded MIMO systems. We first propose a method to generate the symbol synchro sets through the concept of frozen symbols, then refine the symbol synchro sets based on the characteristics analysis of the channel matrix. We optimize the enumerating order of the symbols and reduce the enumerating levels. The frame error rate (FER) and the bit error rate of the proposed algorithms are significantly improved especially for the low-rate codes. The proposed SA-BFSD JDD achieves an up to 7.8 dB performance gain over BFSD at FER
A Synchro-Set-Aided Breadth-First Sphere Decoder for Polar-Coded MIMO Systems
Huayi Zhou
Xiangyun Deng
Yiqian Cai
Yifei Shen
Minhua Yang
X. You
Chuan Zhang
The joint optimization of multiple-input-multiple-output (MIMO) detection and polar decoding has become a research hotspot for future commun… (voir plus)ication systems. The error-correction performance of the separate detection and decoding (SDD) is far from the Shannon capacity, which cannot meet the requirements of communication scenarios such as ultra-reliable and low latency communications (URLLC). The existing joint detection and decoding (JDD) using breadth-first sphere decoding (BFSD) improves the reliability over SDD but still has a huge performance loss on low-rate codes. In this paper, JDD using synchro-set-aided BFSD (SA-BFSD) is proposed to greatly improve the error-correction performance for polar-coded MIMO systems. We first propose a method to generate the symbol synchro sets through the concept of frozen symbols, then refine the symbol synchro sets based on the characteristics analysis of the channel matrix. We optimize the enumerating order of the symbols and reduce the enumerating levels. The frame error rate (FER) and the bit error rate of the proposed algorithms are significantly improved especially for the low-rate codes. The proposed SA-BFSD JDD achieves an up to 7.8 dB performance gain over BFSD at FER