Emergence of deep neural networks (DNNs) has raised enormous attention towards artificial neural networks (ANNs) once again. They have become the state-of-the-art models and have won different machine learning challenges. Although these networks are inspired by the brain, they lack biological plausibility, and they have structural differences compared to the brain. Spiking neural networks (SNNs) have been around for a long time, and they have been investigated to understand the dynamics of the brain. However, their application in real-world and complicated machine learning tasks were limited. Recently, they have shown great potential in solving such tasks. Due to their energy efficiency and temporal dynamics there are many promises in their future development. In this work, we reviewed the structures and performances of SNNs on image classification tasks. The comparisons illustrate that these networks show great capabilities for more complicated problems. Furthermore, the simple learning rules developed for SNNs, such as STDP and R-STDP, can be a potential alternative to replace the backpropagation algorithm used in DNNs.
Language-conditioned policies allow robots to interpret and execute human instructions. Learning such policies requires a substantial investment with regards to time and compute resources. Still, the resulting controllers are highly device-specific and cannot easily be transferred to a robot with different morphology, capability, appearance or dynamics. In this paper, we propose a sample-efficient approach for training language-conditioned manipulation policies that allows for rapid transfer across different types of robots. By introducing a novel method, namely Hierarchical Modularity, and adopting supervised attention across multiple sub-modules, we bridge the divide between modular and end-to-end learning and enable the reuse of functional building blocks. In both simulated and real world robot manipulation experiments, we demonstrate that our method outperforms the current state-of-the-art methods and can transfer policies across 4 different robots in a sample-efficient manner. Finally, we show that the functionality of learned sub-modules is maintained beyond the training process and can be used to introspect the robot decision-making process. Code is available at https://github.com/ir-lab/ModAttn.
Timely and effective feedback within surgical training plays a critical role in developing the skills required to perform safe and efficient surgery. Feedback from expert surgeons, while especially valuable in this regard, is challenging to acquire due to their typically busy schedules, and may be subject to biases. Formal assessment procedures like OSATS and GEARS attempt to provide objective measures of skill, but remain time-consuming. With advances in machine learning there is an opportunity for fast and objective automated feedback on technical skills. The SimSurgSkill 2021 challenge (hosted as a sub-challenge of EndoVis at MICCAI 2021) aimed to promote and foster work in this endeavor. Using virtual reality (VR) surgical tasks, competitors were tasked with localizing instruments and predicting surgical skill. Here we summarize the winning approaches and how they performed. Using this publicly available dataset and results as a springboard, future work may enable more efficient training of surgeons with advances in surgical data science. The dataset can be accessed from https://console.cloud.google.com/storage/browser/isi-simsurgskill-2021.
Contrastive learning methods based on InfoNCE loss are popular in node representation learning tasks on graph-structured data. However, its reliance on data augmentation and its quadratic computational complexity might lead to inconsistency and inefficiency problems. To mitigate these limitations, in this paper, we introduce a simple yet effective contrastive model named Localized Graph Contrastive Learning (Local-GCL in short). Local-GCL consists of two key designs: 1) We fabricate the positive examples for each node directly using its first-order neighbors, which frees our method from the reliance on carefully-designed graph augmentations; 2) To improve the efficiency of contrastive learning on graphs, we devise a kernelized contrastive loss, which could be approximately computed in linear time and space complexity with respect to the graph size. We provide theoretical analysis to justify the effectiveness and rationality of the proposed methods. Experiments on various datasets with different scales and properties demonstrate that in spite of its simplicity, Local-GCL achieves quite competitive performance in self-supervised node representation learning tasks on graphs with various scales and properties.
For scalable coding, a high quality of the lowpass band of a wavelet transform is crucial when it is used as a downscaled version of the original signal. However, blur and motion can lead to disturbing artifacts. By incorporating feasible compensation methods directly into the wavelet transform, the quality of the lowpass band can be improved. The displacement in dynamic medical 3-D+t volumes from Computed Tomography is mainly given by expansion and compression of tissue over time and can be modeled well by mesh-based methods. We extend a 2-D mesh-based compensation method to three dimensions to obtain a volume compensation method that can additionally compensate deforming displacements in the third dimension. We show that a 3-D mesh can obtain a higher quality of the lowpass band by 0.28 dB with less than 40% of the model parameters of a comparable 2-D mesh. Results from lossless coding with JPEG 2000 3D and SPECK3D show that the compensated subbands using a 3-D mesh need about 6% less data compared to using a 2-D mesh.
Vehicle re-identification (Re-ID) is a critical component of the autonomous driving perception system, and research in this area has accelerated in recent years. However, there is yet no perfect solution to the vehicle re-identification issue associated with the car's surround-view camera system. Our analysis identifies two significant issues in the aforementioned scenario: i) It is difficult to identify the same vehicle in many picture frames due to the unique construction of the fisheye camera. ii) The appearance of the same vehicle when seen via the surround vision system's several cameras is rather different. To overcome these issues, we suggest an integrative vehicle Re-ID solution method. On the one hand, we provide a technique for determining the consistency of the tracking box drift with respect to the target. On the other hand, we combine a Re-ID network based on the attention mechanism with spatial limitations to increase performance in situations involving multiple cameras. Finally, our approach combines state-of-the-art accuracy with real-time performance. We will soon make the source code and annotated fisheye dataset available.
Probabilistic user modeling is essential for building collaborative AI systems within probabilistic frameworks. However, modern advanced user models, often designed as cognitive behavior simulators, are computationally prohibitive for interactive use in cooperative AI assistants. In this extended abstract, we address this problem by introducing widely-applicable differentiable surrogates for bypassing this computational bottleneck; the surrogates enable using modern behavioral models with online computational cost which is independent of their original computational cost. We show experimentally that modeling capabilities comparable to likelihood-free inference methods are achievable, with over eight orders of magnitude reduction in computational time. Finally, we demonstrate how AI-assistants can computationally feasibly use cognitive models in a previously studied menu-search task.
The two popular datasets ScanRefer [16] and ReferIt3D [3] connect natural language to real-world 3D data. In this paper, we curate a large-scale and complementary dataset extending both the aforementioned ones by associating all objects mentioned in a referential sentence to their underlying instances inside a 3D scene. Specifically, our Scan Entities in 3D (ScanEnts3D) dataset provides explicit correspondences between 369k objects across 84k natural referential sentences, covering 705 real-world scenes. Crucially, we show that by incorporating intuitive losses that enable learning from this novel dataset, we can significantly improve the performance of several recently introduced neural listening architectures, including improving the SoTA in both the Nr3D and ScanRefer benchmarks by 4.3% and 5.0%, respectively. Moreover, we experiment with competitive baselines and recent methods for the task of language generation and show that, as with neural listeners, 3D neural speakers can also noticeably benefit by training with ScanEnts3D, including improving the SoTA by 13.2 CIDEr points on the Nr3D benchmark. Overall, our carefully conducted experimental studies strongly support the conclusion that, by learning on ScanEnts3D, commonly used visio-linguistic 3D architectures can become more efficient and interpretable in their generalization without needing to provide these newly collected annotations at test time. The project's webpage is https://scanents3d.github.io/ .
We present RAVEn, a self-supervised multi-modal approach to jointly learn visual and auditory speech representations. Our pre-training objective involves encoding masked inputs, and then predicting contextualised targets generated by slowly-evolving momentum encoders. Driven by the inherent differences between video and audio, our design is asymmetric w.r.t. the two modalities' pretext tasks: Whereas the auditory stream predicts both the visual and auditory targets, the visual one predicts only the auditory targets. We observe strong results in low- and high-resource labelled data settings when fine-tuning the visual and auditory encoders resulting from a single pre-training stage, in which the encoders are jointly trained. Notably, RAVEn surpasses all self-supervised methods on visual speech recognition (VSR) on LRS3, and combining RAVEn with self-training using only 30 hours of labelled data even outperforms a recent semi-supervised method trained on 90,000 hours of non-public data. At the same time, we achieve state-of-the-art results in the LRS3 low-resource setting for auditory speech recognition (as well as for VSR). Our findings point to the viability of learning powerful speech representations entirely from raw video and audio, i.e., without relying on handcrafted features. Code and models will be made public.
Contemporary Virtual Reality (VR) setups often include an external source delivering content to a Head-Mounted Display (HMD). "Cutting the wire" in such setups and going truly wireless will require a wireless network capable of delivering enormous amounts of video data at an extremely low latency. The massive bandwidth of higher frequencies, such as the millimeter-wave (mmWave) band, can meet these requirements. Due to high attenuation and path loss in the mmWave frequencies, beamforming is essential. In wireless VR, where the antenna is integrated into the HMD, any head rotation also changes the antenna's orientation. As such, beamforming must adapt, in real-time, to the user's head rotations. An HMD's built-in sensors providing accurate orientation estimates may facilitate such rapid beamforming. In this work, we present coVRage, a receive-side beamforming solution tailored for VR HMDs. Using built-in orientation prediction present on modern HMDs, the algorithm estimates how the Angle of Arrival (AoA) at the HMD will change in the near future, and covers this AoA trajectory with a dynamically shaped oblong beam, synthesized using sub-arrays. We show that this solution can cover these trajectories with consistently high gain, even in light of temporally or spatially inaccurate orientational data.