The key challenge of multi-view indoor 3D object detection is to infer accurate geometry information from images for precise 3D detection. Previous method relies on NeRF for geometry reasoning. However, the geometry extracted from NeRF is generally inaccurate, which leads to sub-optimal detection performance. In this paper, we propose MVSDet which utilizes plane sweep for geometry-aware 3D object detection. To circumvent the requirement for a large number of depth planes for accurate depth prediction, we design a probabilistic sampling and soft weighting mechanism to decide the placement of pixel features on the 3D volume. We select multiple locations that score top in the probability volume for each pixel and use their probability score to indicate the confidence. We further apply recent pixel-aligned Gaussian Splatting to regularize depth prediction and improve detection performance with little computation overhead. Extensive experiments on ScanNet and ARKitScenes datasets are conducted to show the superiority of our model.

We propose a new algorithm for model-based distributional reinforcement learning (RL), and prove that it is minimax-optimal for approximating return distributions in the generative model regime (up to logarithmic factors), the first result of this kind for any distributional RL algorithm. Our analysis also provides new theoretical perspectives on categorical approaches to distributional RL, as well as introducing a new distributional Bellman equation, the stochastic categorical CDF Bellman equation, which we expect to be of independent interest. Finally, we provide an experimental study comparing a variety of model-based distributional RL algorithms, with several key takeaways for practitioners.

For the first time, our research introduces continual learning for abstract logical concepts, which mimics the process of humans acquiring higher-order learning abilities. This moves away from the traditional CL approach centered around images and considers a new direction that CL should ultimately strive for. Algorithmic reasoning tasks are fundamentally different from image data in their abstract and logical nature. The discontinuity of input data, the necessity for generalization regarding out-of-distribution samples, and the inability to use data augmentation or mix-up techniques present the need for new CL algorithms that differ from existing methodologies. We hope that future research will further explore methodologies that effectively leverage these unique characteristics of AR.

Continual learning (CL) aims to incrementally learn multiple tasks that are presented sequentially. The significance of CL lies not only in the practical importance but also in studying the learning mechanisms of humans who are excellent continual learners. While most research on CL has been done on structured data such as images, there is a lack of research on CL for abstract logical concepts such as counting, sorting, and arithmetic, which humans learn gradually over time in the real world. In this work, for the first time, we introduce novel algorithmic reasoning (AR) methodology for continual tasks of abstract concepts: CLeAR. Our methodology proposes a one-to-many mapping of input distribution to a shared mapping space, which allows the alignment of various tasks of different dimensions and shared semantics. Our tasks of abstract logical concepts, in the form of formal language, can be classified into Chomsky hierarchies based on their difficulty. In this study, we conducted extensive experiments consisting of 15 tasks with various levels of Chomsky hierarchy, ranging from in-hierarchy to inter-hierarchy scenarios. CLeAR not only achieved near zero forgetting but also improved accuracy during following tasks, a phenomenon known as backward transfer, while previous CL methods designed for image classification drastically failed.

In many machine learning tasks, data is inherently sequential. Most existing algorithms learn from sequential data in an auto-regressive manner, which predicts the next unseen data point based on the observed sequence, implicitly assuming the presence of an evolving pattern embedded in the data that can be leveraged. However, identifying and assessing evolving patterns in learning tasks heavily relies on human expertise, and lacks a standardized quantitative measure. In this paper, we show that such a measure enables us to determine the suitability of employing sequential models, measure the temporal order of time series data, and conduct feature/data selections, which can be beneficial to a variety of learning tasks: time-series forecastings, classification tasks with temporal distribution shift, video predictions, etc.

The recursion of this case is also the same (see Eqn. 4): D

In current deep learning tasks, Adam-style optimizers--such as Adam, Adagrad, RMSprop, Adafactor, and Lion--have been widely used as alternatives to SGDstyle optimizers. These optimizers typically update model parameters using the sign of gradients, resulting in more stable convergence curves. The learning rate and the batch size are the most critical hyperparameters for optimizers, which require careful tuning to enable effective convergence. Previous research has shown that the optimal learning rate increases linearly (or follows similar rules) with batch size for SGD-style optimizers. However, this conclusion is not applicable to Adam-style optimizers.

We develop EigenVI, an eigenvalue-based approach for black-box variational inference (BBVI).