اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدموديول الشبكة الشامل للتفكير البصري التركيبي
Meta Module Network for Compositional Visual Reasoning
761
0
0.0
(
0
)
اسأل ChatGPT حول البحث
تتميز الشبكة الوحدات العصبية (NMN) بالفهم القوي والتركيبية بفضل وحدات العصبية المصنوعة بشكل يدوي مع القدرة على التفكير بالمسافات المتعددة. ومع ذلك، يعاني معظم NMNs من عيوبان حاسمتان: 1) التكبير: تجعل الوحدة المخصصة للوظيفة مشوشا عند التكبير إلى مجموعة أكبر من الوظائف في المهام المعقدة؛ 2) التطبيقية: تجعل سجل الوحدات المحدد مسبقا بشكل صلب من الصعب التطبيق على الوظائف الغير مرئية في مهام / دول مختلفة. لتصميم بنية شبكة NMN أقوى للاستخدام العملي، نقترح Meta Module Network (MMN) محورة على وحدة التحليل المتطورة، والتي يمكنها استيعاب وصفات الوظيفة وتحول إلى وحدات مختلفة بشكل ديناميكي. ثم تتم دمج الوحدات الفردية في برنامج تنفيذي للتفكير البصري المعقد، والذي يورث الفهم القوي والتركيبية ل NMN. مع هذا الآلية التطبيقية المرنة، يتم ارث المعلمات من الوحدة المتطورة المركزية للوحدات الفردية، والحفاظ على نفس معدل الصعوبة للنموذج عندما ينمو مجموعة الوظائف، مما يعد بأمانة أفضل للتكبير. في الوقت نفسه، حيث تم ترميز الوظائف في مساحة التضمين، يمكن تمثيل الوظائف الغير مرئية بسهولة على أساس تشابهها الهيكلي مع الوظائف المرئية مسبقا، مما يضمن أفضل تطبيقية. تأكد التجارب على بيانات GQA و CLEVR من التفوق الذي تمتلكه MMN على تصميمات NMN الحديثة. كما تثبت التجارب الصناعية على الوظائف الغير مرئية المحجوزة من بيانات GQA قوة التطبيقية القوية ل MMN. تم إصدار الرمز والنموذج في Github https://github.com/wenhuchen/Meta-Module-Network.
Neural Module Network (NMN) exhibits strong interpretability and compositionality thanks to its handcrafted neural modules with explicit multi-hop reasoning capability. However, most NMNs suffer from two critical drawbacks: 1) scalability: customized module for specific function renders it impractical when scaling up to a larger set of functions in complex tasks; 2) generalizability: rigid pre-defined module inventory makes it difficult to generalize to unseen functions in new tasks/domains. To design a more powerful NMN architecture for practical use, we propose Meta Module Network (MMN) centered on a novel meta module, which can take in function recipes and morph into diverse instance modules dynamically. The instance modules are then woven into an execution graph for complex visual reasoning, inheriting the strong explainability and compositionality of NMN. With such a flexible instantiation mechanism, the parameters of instance modules are inherited from the central meta module, retaining the same model complexity as the function set grows, which promises better scalability. Meanwhile, as functions are encoded into the embedding space, unseen functions can be readily represented based on its structural similarity with previously observed ones, which ensures better generalizability. Experiments on GQA and CLEVR datasets validate the superiority of MMN over state-of-the-art NMN designs. Synthetic experiments on held-out unseen functions from GQA dataset also demonstrate the strong generalizability of MMN. Our code and model are released in Github https://github.com/wenhuchen/Meta-Module-Network.
قيم البحث
اقرأ أيضاً
Humans learn to solve tasks of increasing complexity by building on top of previously acquired knowledge. Typically, there exists a natural progression in the tasks that we learn - most do not require completely independent solutions, but can be brok
en down into simpler subtasks. We propose to represent a solver for each task as a neural module that calls existing modules (solvers for simpler tasks) in a functional program-like manner. Lower modules are a black box to the calling module, and communicate only via a query and an output. Thus, a module for a new task learns to query existing modules and composes their outputs in order to produce its own output. Our model effectively combines previous skill-sets, does not suffer from forgetting, and is fully differentiable. We test our model in learning a set of visual reasoning tasks, and demonstrate improved performances in all tasks by learning progressively. By evaluating the reasoning process using human judges, we show that our model is more interpretable than an attention-based baseline.
In this paper, we propose the Broadcasting Convolutional Network (BCN) that extracts key object features from the global field of an entire input image and recognizes their relationship with local features. BCN is a simple network module that collect
s effective spatial features, embeds location information and broadcasts them to the entire feature maps. We further introduce the Multi-Relational Network (multiRN) that improves the existing Relation Network (RN) by utilizing the BCN module. In pixel-based relation reasoning problems, with the help of BCN, multiRN extends the concept of `pairwise relations in conventional RNs to `multiwise relations by relating each object with multiple objects at once. This yields in O(n) complexity for n objects, which is a vast computational gain from RNs that take O(n^2). Through experiments, multiRN has achieved a state-of-the-art performance on CLEVR dataset, which proves the usability of BCN on relation reasoning problems.
Abstract reasoning refers to the ability to analyze information, discover rules at an intangible level, and solve problems in innovative ways. Ravens Progressive Matrices (RPM) test is typically used to examine the capability of abstract reasoning. T
he subject is asked to identify the correct choice from the answer set to fill the missing panel at the bottom right of RPM (e.g., a 3$times$3 matrix), following the underlying rules inside the matrix. Recent studies, taking advantage of Convolutional Neural Networks (CNNs), have achieved encouraging progress to accomplish the RPM test. However, they partly ignore necessary inductive biases of RPM solver, such as order sensitivity within each row/column and incremental rule induction. To address this problem, in this paper we propose a Stratified Rule-Aware Network (SRAN) to generate the rule embeddings for two input sequences. Our SRAN learns multiple granularity rule embeddings at different levels, and incrementally integrates the stratified embedding flows through a gated fusion module. With the help of embeddings, a rule similarity metric is applied to guarantee that SRAN can not only be trained using a tuplet loss but also infer the best answer efficiently. We further point out the severe defects existing in the popular RAVEN dataset for RPM test, which prevent from the fair evaluation of the abstract reasoning ability. To fix the defects, we propose an answer set generation algorithm called Attribute Bisection Tree (ABT), forming an improved dataset named Impartial-RAVEN (I-RAVEN for short). Extensive experiments are conducted on both PGM and I-RAVEN datasets, showing that our SRAN outperforms the state-of-the-art models by a considerable margin.
Recently, studies of visual question answering have explored various architectures of end-to-end networks and achieved promising results on both natural and synthetic datasets, which require explicitly compositional reasoning. However, it has been ar
gued that these black-box approaches lack interpretability of results, and thus cannot perform well on generalization tasks due to overfitting the dataset bias. In this work, we aim to combine the benefits of both sides and overcome their limitations to achieve an end-to-end interpretable structural reasoning for general images without the requirement of layout annotations. Inspired by the property of a capsule network that can carve a tree structure inside a regular convolutional neural network (CNN), we propose a hierarchical compositional reasoning model called the Linguistically driven Graph Capsule Network, where the compositional process is guided by the linguistic parse tree. Specifically, we bind each capsule in the lowest layer to bridge the linguistic embedding of a single word in the original question with visual evidence and then route them to the same capsule if they are siblings in the parse tree. This compositional process is achieved by performing inference on a linguistically driven conditional random field (CRF) and is performed across multiple graph capsule layers, which results in a compositional reasoning process inside a CNN. Experiments on the CLEVR dataset, CLEVR compositional generation test, and FigureQA dataset demonstrate the effectiveness and composition generalization ability of our end-to-end model.
Reasoning about images/objects and their hierarchical interactions is a key concept for the next generation of computer vision approaches. Here we present a new framework to deal with it through a visual hierarchical context-based reasoning. Current
reasoning methods use the fine-grained labels from images objects and their interactions to predict labels to new objects. Our framework modifies this current information flow. It goes beyond and is independent of the fine-grained labels from the objects to define the image context. It takes into account the hierarchical interactions between different abstraction levels (i.e. taxonomy) of information in the images and their bounding-boxes. Besides these connections, it considers their intrinsic characteristics. To do so, we build and apply graphs to graph convolution networks with convolutional neural networks. We show a strong effectiveness over widely used convolutional neural networks, reaching a gain 3 times greater on well-known image datasets. We evaluate the capability and the behavior of our framework under different scenarios, considering distinct (superclass, subclass and hierarchical) granularity levels. We also explore attention mechanisms through graph attention networks and pre-processing methods considering dimensionality expansion and/or reduction of the features representations. Further analyses are performed comparing supervised and semi-supervised approaches.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
