3D scene understanding from point clouds plays a vital role for various robotic applications. Unfortunately, current state-of-the-art methods use separate neural networks for different tasks like object detection or room layout estimation. Such a sch
eme has two limitations: 1) Storing and running several networks for different tasks are expensive for typical robotic platforms. 2) The intrinsic structure of separate outputs are ignored and potentially violated. To this end, we propose the first transformer architecture that predicts 3D objects and layouts simultaneously, using point cloud inputs. Unlike existing methods that either estimate layout keypoints or edges, we directly parameterize room layout as a set of quads. As such, the proposed architecture is termed as P(oint)Q(uad)-Transformer. Along with the novel quad representation, we propose a tailored physical constraint loss function that discourages object-layout interference. The quantitative and qualitative evaluations on the public benchmark ScanNet show that the proposed PQ-Transformer succeeds to jointly parse 3D objects and layouts, running at a quasi-real-time (8.91 FPS) rate without efficiency-oriented optimization. Moreover, the new physical constraint loss can improve strong baselines, and the F1-score of the room layout is significantly promoted from 37.9% to 57.9%.
Aspect Sentiment Triplet Extraction (ASTE) aims to extract triplets from sentences, where each triplet includes an entity, its associated sentiment, and the opinion span explaining the reason for the sentiment. Most existing research addresses this p
roblem in a multi-stage pipeline manner, which neglects the mutual information between such three elements and has the problem of error propagation. In this paper, we propose a Semantic and Syntactic Enhanced aspect Sentiment triplet Extraction model (S3E2) to fully exploit the syntactic and semantic relationships between the triplet elements and jointly extract them. Specifically, we design a Graph-Sequence duel representation and modeling paradigm for the task of ASTE: we represent the semantic and syntactic relationships between word pairs in a sentence by graph and encode it by Graph Neural Networks (GNNs), as well as modeling the original sentence by LSTM to preserve the sequential information. Under this setting, we further apply a more efficient inference strategy for the extraction of triplets. Extensive evaluations on four benchmark datasets show that S3E2 significantly outperforms existing approaches, which proves our S3E2s superiority and flexibility in an end-to-end fashion.
We address the degree to which machine learning can be used to accurately and transferably predict post-Hartree-Fock correlation energies. Refined strategies for feature design and selection are presented, and the molecular-orbital-based machine lear
ning (MOB-ML) method is applied to several test systems. Strikingly, for the MP2, CCSD, and CCSD(T) levels of theory, it is shown that the thermally accessible (350 K) potential energy surface for a single water molecule can be described to within 1 millihartree using a model that is trained from only a single reference calculation at a randomized geometry. To explore the breadth of chemical diversity that can be described, MOB-ML is also applied to a new dataset of thermalized (350 K) geometries of 7211 organic models with up to seven heavy atoms. In comparison with the previously reported $Delta$-ML method, MOB-ML is shown to reach chemical accuracy with three-fold fewer training geometries. Finally, a transferability test in which models trained for seven-heavy-atom systems are used to predict energies for thirteen-heavy-atom systems reveals that MOB-ML reaches chemical accuracy with 36-fold fewer training calculations than $Delta$-ML (140 versus 5000 training calculations).
We investigate the communication performance of a few-mode EDFA based all-optical relaying system for atmospheric channels in this paper. A dual-hop free space optical communication model based on the relay with two-mode EDFA is derived. The BER perf
ormance is numerically calculated. Compared with all-optical relaying system with single-mode EDFA, the power budget is increased by 4 dB, 7.5 dB and 11.5 dB at BER = 1E-4 under the refractive index structure constant Cn2 = 2E-14, 5E-14 and 1E-13 respectively when a few mode fiber supporting 4 modes is utilized as the receiving fiber at the destination. The optimal relay location is slightly backward from the middle of the link. The BER performance is the best when mode-dependent gain of FM-EDFA is zero.
Within the mean-field theory, we investigate the magnetic properties of a charged spin-1 Bose gas in two dimension. In this system the diamagnetism competes with paramagnetism, where Lande-factor $g$ is introduced to describe the strength of the para
magnetic effect. The system presents a crossover from diamagnetism to paramagnetism with the increasing of Lande-factor. The critical value of the Lande-factor, $g_{c}$, is discussed as a function of the temperature and magnetic field. We get the same value of $g_{c}$ both in the low temperature and strong magnetic field limit. Our results also show that in very weak magnetic field no condensation happens in the two dimensional charged spin-1 Bose gas.
