United States

There are two key issues that limit further improvements in the performance of existing rotational detectors: 1) Periodic sudden change of the parameters in the rotating bounding box (RBBox) definition causes a numerical discontinuity in the loss (such as smoothL1 loss). 2) There is a gap of optimization asynchrony between the loss in the RBBox regression and evaluation metrics. In this paper, we define a new distance formulation between two convex polygons describing the overlapping degree and non-overlapping degree. Based on this smooth distance, we propose a loss called Polygon-to-Polygon distance loss (P2P Loss). The distance is derived from the area sum of triangles specified by the vertexes of one polygon and the edges of the other. Therefore, the P2P Loss is continuous, differentiable, and inherently free from any RBBox definition. Our P2P Loss is not only consistent with the detection metrics but also able to measure how far, as well as how similar, a RBBox is from another one even when they are completely non-overlapping. These features allow the RetinaNet using the P2P Loss to achieve 79.15% mAP on the DOTA dataset, which is quite competitive compared with many state-of-the-art rotated object detectors.

AAAI 2022

Polygon-to-Polygon Distance Loss for Rotated Object Detection

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2021

technical paper

Fine-grained visual classification (FGVC) which aims at recognizing objects from subcategories is a very challenging task due to the inherently subtle inter-class differences. Most existing works mainly tackle this problem by reusing the backbone network to extract features of detected discriminative regions. However, this strategy inevitably complicates the pipeline and pushes the proposed regions to contain most parts of the objects thus fails to locate the really important parts. Recently, vision transformer (ViT) shows its strong performance in the traditional classification task. The self-attention mechanism of the transformer links every patch token to the classification token. In this work, we first evaluate the effectiveness of the ViT framework in the fine-grained recognition setting. Then motivated by the strength of the attention link can be intuitively considered as an indicator of the importance of tokens, we further propose a novel Part Selection Module that can be applied to most of the transformer architectures where we integrate all raw attention weights of the transformer into an attention map for guiding the network to effectively and accurately select discriminative image patches and compute their relations. A contrastive loss is applied to enlarge the distance between feature representations of confusing classes. We name the augmented transformer-based model TransFG and demonstrate the value of it by conducting experiments on five popular fine-grained benchmarks where we achieve state-of-the-art performance. Qualitative results are presented for better understanding of our model.

TransFG: A Transformer Architecture for Fine-Grained Recognition

The paper proposes a probabilistic generalization of the well-known Strong Backdoor Set (SBS) concept applied to the Boolean Satisfiability Problem (SAT). We call a set of Boolean variables B a Ï-backdoor, if for a fraction of at least Ï of possible assignments of variables from B, assigning their values to variables in a Boolean formula in Conjunctive Normal Form (CNF) results in polynomially solvable formulas. Clearly, a Ï-backdoor with Ï=1 is an SBS. For a given set B it is possible to efficiently construct an (Îµ, Î´)-approximation of parameter Ï using the Monte Carlo method. Thus, we define an (Îµ, Î´)-SBS as such a set B for which the conclusion "parameter Ï deviates from 1 by no more than Îµ" is true with probability no smaller than 1 - Î´. We consider the problems of finding the minimum SBS and the minimum (Îµ, Î´)-SBS. To solve the former problem, one can use the algorithm described by R. Williams, C. Gomes and B. Selman in 2003. In the paper we propose a new probabilistic algorithm to solve the latter problem, and show that the asymptotic estimation of the worst-case complexity of the proposed algorithm is significantly smaller than that of the algorithm by Williams et al. For practical applications, we suggest a metaheuristic optimization algorithm based on the penalty function method to seek the minimal (Îµ, Î´)-SBS. Results of computational experiments show that the use of (Îµ, Î´)-SBSes found by the proposed algorithm allows speeding up solving of test problems related to equivalence checking and hard crafted and combinatorial benchmarks compared to state-of-the-art SAT solvers.

On Probabilistic Generalization of Backdoors in Boolean Satisfiability

Inferring spatial relations in natural language is a crucial ability an intelligent system should possess. The bAbI dataset tries to capture tasks relevant to this domain (task 17 and 19). However, these tasks have several limitations. Most importantly, they are limited to fixed expressions, they are limited in the number of reasoning steps required to solve them, and they fail to test the robustness of models to input that contains irrelevant or redundant information. In this paper, we present a new Question-Answering dataset called StepGame for robust multi-step spatial reasoning in texts. Our experiments demonstrate that state-of-the-art models on the bAbI dataset struggle on the StepGame dataset. Moreover, we propose a Tensor-Product based Memory-Augmented Neural Network (TP-MANN) specialized for spatial reasoning tasks. Experimental results on both datasets show that our model outperforms all the baselines with superior generalization and robustness performance.

StepGame: A New Benchmark for Robust Multi-Hop Spatial Reasoning in Texts

Learning effective representations of entities and relations for knowledge graphs (KGs) is critical to the success of many multi-relational learning tasks. Existing methods based on graph neural networks learn a deterministic embedding function, which lacks sufficient flexibility to explore better choices when dealing with the imperfect and noisy KGs such as the scarce labeled nodes and noisy graph structure. To this end, we propose a novel multi-relational graph Gaussian Process network (GGPN), which aims to improve the flexibility of deterministic methods by simultaneously learning a family of embedding functions, i.e., a stochastic embedding function. Specifically, a Bayesian Gaussian Process (GP) is proposed to model the distribution of this stochastic function and the resulting representations are obtained by aggregating stochastic function values, i.e., messages, from neighboring entities. The two problems incurred when leveraging GP in GGPN are the proper choice of kernel function and the cubic computational complexity. To address the first problem, we further propose a novel kernel function that can explicitly take the diverse relations between each pair of entities into account and be adaptively learned in a data-driven way. We address the second problem by reformulating GP as a Bayesian linear model, resulting in a linear computational complexity. With these two solutions, our GGPN can be efficiently trained in an end-to-end manner. We evaluate our GGPN in link prediction and entity classification tasks, and the experimental results demonstrate the superiority of our method. Our code is available at https://github.com/sysu-gzchen/GGPN.

Multi-Relational Graph Representation Learning with Bayesian Gaussian Process Network

Color Constancy aims to correct image color casts caused by scene illumination. Recently, although the deep learning approaches have remarkably improved on single-camera data, these models still suffer from the seriously insufficient data problem, resulting in shallow model capacity and degradation in multi-camera settings. In this paper, to alleviate this problem, we present a Transfer Learning Color Constancy (TLCC) method that leverages cross-camera RAW data and massive unlabeled sRGB data to support training. Specifically, TLCC consists of the Statistic Estimation Scheme (SE-Scheme) and Color-Guided Adaption Branch (CGA-Branch). SE-Scheme builds a statistic perspective to map the camera-related illumination labels into camera-agnostic form and produce pseudo labels for sRGB data, which greatly expands data for joint training. Then, CGA-Branch further promotes efficient transfer learning from sRGB to RAW data by extracting color information to regularize the backbone's features adaptively. Experimental results show the TLCC has overcome the data limitation and model degradation, outperforming the state-of-the-art performance on popular benchmarks. Moreover, the experiments also prove the TLCC is capable of learning new scenes information from sRGB data to improve accuracy on the RAW images with similar scenes.

Transfer Learning for Color Constancy via Statistic Perspective

This paper presents a model-free reinforcement learning (RL) algorithm for infinite-horizon average-reward Constrained Markov Decision Processes (CMDPs). Considering a learning horizon K, which is sufficiently large, the proposed algorithm achieves sublinear regret and zero constraint violation. The bounds depend on the number of states S, the number of actions A, and two constants which are independent of the learning horizon K.

A Provably-Efficient Model-Free Algorithm for Infinite-Horizon Average-Reward Constrained Markov Decision Processes

Recent studies have shown that strong Natural Language Understanding (NLU) models are prone to relying on annotation biases of the datasets as a shortcut, which goes against the underlying mechanisms of the task of interest. To reduce such biases, several recent works introduce debiasing methods to regularize the training process of targeted NLU models. In this paper, we provide a new perspective with causal inference to find out the bias. On one hand, we show that there is an unobserved confounder for the natural language utterances and their respective classes, leading to spurious correlations from training data. To remove such confounder, the backdoor adjustment with causal intervention is utilized to find the true causal effect, which makes the training process fundamentally different from the traditional likelihood estimation. On the other hand, in inference process, we formulate the bias as the direct causal effect and remove it by pursuing the indirect causal effect with counterfactual reasoning. We conduct experiments on large-scale natural language inference and fact verification benchmarks, evaluating on bias sensitive datasets that are specifically designed to assess the robustness of models against known biases in the training data. Experimental results show that our proposed debiasing framework outperforms previous state-of-the-art debiasing methods while maintaining the original in-distribution performance.

Debiasing NLU Models via Causal Intervention and Counterfactual Reasoning

We study reward design strategies for incentivizing a reinforcement learning agent to adopt a policy from a set of admissible policies. The goal of the reward designer is to modify the underlying reward function cost-efficiently while ensuring that any approximately optimal deterministic policy under the new reward function is admissible and performs well under the original reward function. This problem can be viewed as a dual to the problem of optimal reward poisoning attacks: instead of forcing an agent to adopt a specific policy, the reward designer incentivizes an agent to avoid taking actions that are inadmissible in certain states. Perhaps surprisingly, and in contrast to the problem of optimal reward poisoning attacks, we first show that the reward design problem for admissible policy teaching is computationally challenging, and it is NP-hard to find an approximately optimal reward modification. We then proceed by formulating a surrogate problem whose optimal solution approximates the optimal solution to the reward design problem in our setting, but is more amenable to optimization techniques and analysis. For this surrogate problem, we present characterization results that provide bounds on the value of the optimal solution. Finally, we design a local search algorithm to solve the surrogate problem and showcase its utility using simulation-based experiments.

Admissible Policy Teaching through Reward Design

Unlike traditional supervised learning, in many settings only partial feedback is available. We may only observe outcomes for the chosen actions, but not the counterfactual outcomes associated with other alternatives. Such settings encompass a wide variety of applications including pricing, online marketing and precision medicine. A key challenge is that observational data are influenced by historical policies deployed in the system, yielding a biased data distribution. We approach this task as a domain adaptation problem and propose a self-training algorithm which imputes outcomes with categorical values for finite unseen actions in the observational data to simulate a randomized trial through pseudolabelling, which we refer to as Counterfactual Self-Training (CST). CST iteratively imputes pseudolabels and retrains the model. In addition, we show input consistency loss can further improve CST performance which is shown in recent theoretical analysis of pseudolabelling. We demonstrate the effectiveness of the proposed algorithms on both synthetic and real datasets.

Enhancing Counterfactual Classification Performance via Self-Training

Data augmentation has been an important ingredient for boosting performances of learned models. Prior data augmentation methods for few-shot text classification have led to great performance boosts. However, they have not been designed to capture the intricate compositional structure of natural language. As a result, they fail to generate samples with plausible and diverse sentence structures. Motivated by this, we present the data Augmentation using Lexicalized Probabilistic context-free grammars (ALP) that generates augmented samples with diverse syntactic structures with plausible grammar. The lexicalized PCFG parse trees consider both the constituents and dependencies to produce a syntactic frame that maximizes a variety of word choices in a syntactically preservable manner without specific domain experts. Experiments on few-shot text classification tasks demonstrate that ALP enhances many state-of-the-art classification methods. As a second contribution, we delve into the train-val splitting methodologies when a data augmentation method comes into play. We argue empirically that the traditional splitting of training and validation sets is sub-optimal compared to our novel augmentation-based splitting strategies that further expand the training split with the same number of labeled data. Taken together, our contributions on the data augmentation strategies yield a strong training recipe for few-shot text classification tasks.

Downloads

Next from AAAI 2022

TransFG: A Transformer Architecture for Fine-Grained Recognition

Similar lecture

Learning Modulated Loss for Rotated Object Detection

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2022

TransFG: A Transformer Architecture for Fine-Grained Recognition

Similar lecture

Learning Modulated Loss for Rotated Object Detection

Downloads