Recent successes of Deep Neural Networks (DNNs) in a variety of research tasks, however, heavily rely on the large amounts of labeled samples. This may require considerable annotation cost in real-world applications. Fortunately, active learning is a promising methodology to train high-performing model with minimal annotation cost. In the deep learning context, the critical question of active learning is how to precisely identify the informativeness of samples for DNN. In this paper, inspired by piece-wise linear interpretability in DNN, we introduce the linearly separable regions of samples to the problem of active learning, and propose a novel Deep Active learning approach by Model Interpretability (DAMI). To keep the maximal representativeness of the entire unlabeled data, DAMI tries to select and label samples on different linearly separable regions introduced by the piece-wise linear interpretability in DNN. We focus on modeling Multi-Layer Perception (MLP) for modeling tabular data. Specifically, we use the local piece-wise interpretation in MLP as the representation of each sample, and directly run K-Center clustering to select and label samples. To be noted, this whole process of DAMI does not require any hyper-parameters to tune manually. To verify the effectiveness of our approach, extensive experiments have been conducted on several tabular datasets. The experimental results demonstrate that DAMI constantly outperforms several state-of-the-art compared approaches.

We present a causal speech enhancement model working on the raw waveform that runs in real-time on a laptop CPU. The proposed model is based on an encoder-decoder architecture with skip-connections. It is optimized on both time and frequency domains, using multiple loss functions. Empirical evidence shows that it is capable of removing various kinds of background noise including stationary and non-stationary noises, as well as room reverb. Additionally, we suggest a set of data augmentation techniques applied directly on the raw waveform which further improve model performance and its generalization abilities. We perform evaluations on several standard benchmarks, both using objective metrics and human judgements. The proposed model matches state-of-the-art performance of both causal and non causal methods while working directly on the raw waveform.

Deep neural networks (DNNs) have achieved great success in image classification, but they may be very vulnerable to adversarial attacks with small perturbations to images. Moreover, the adversarial training based on adversarial image samples has been shown to improve the robustness and generalization of DNNs. The aim of this paper is to develop a novel framework based on information-geometry sensitivity analysis and the particle swarm optimization to improve two aspects of adversarial image generation and training for DNNs. The first one is customized generation of adversarial examples. It can design adversarial attacks from options of the number of perturbed pixels, the misclassification probability, and the targeted incorrect class, and hence it is more flexible and effective to locate vulnerable pixels and also enjoys certain adversarial universality. The other is targeted adversarial training. DNN models can be improved in training with the adversarial information using a manifold-based influence measure effective in vulnerable image/pixel detection as well as allowing for targeted attacks, thereby exhibiting an enhanced adversarial defense in testing.

Multi-label classification involves predicting zero or more class labels. Unlike normal classification tasks where class labels are mutually exclusive, multi-label classification requires specialized machine learning algorithms that support predicting multiple mutually non-exclusive classes or “labels.” Deep learning neural networks are an example of an algorithm that natively supports multi-label classification problems. Neural network models for […]

Before going through the list of popular Machine Learning frameworks, let's understand why it is important to embrace ML skills. We all know that Machine Learning and Artificial Intelligence are the hot trends in Technology. According to Gardner, these Technologies will create 2.3 million jobs in the future. It has been seen that there are lots of tools and Frameworks that help us to easily implement Machine learning in our application. So let's have a quick introduction to Machine Learning.

Experts claim artificial intelligence (AI) will shape the lives of people within the next ten years. Ultimately, AI can be a game-changer in amplifying human effectiveness and remove mundane tasks. "AI is going to change the world more than anything in the history of mankind. Artificial intelligence is being used by almost all the major industries -- from education to healthcare to retail and e-commerce, public relations (PR), small businesses, recruitment and services, and manufacturing. Forward to 2030, AI is said to contribute USD 15 trillion to the world economy, source PwC.

Late advancements in artificial intelligence have recharged interest in building frameworks that learn and think as individuals. Numerous advances have originated from utilizing deep neural networks trained end-to-end in operations, for example, object recognition, video games, and board games, accomplishing tasks that are equal to or even beats people in certain regards. In spite of their biological inspiration and performance achievements, these frameworks are different from human intelligence in essential ways. Cognitive science is growing and proposing human-like learning and thinking machines should reach past current engineering trends in both what they learn, and how they learn it. It was conceived from pattern recognition and the theory that PCs can learn without being programmed to perform explicit tasks; scientists intrigued by artificial intelligence needed to check whether computers could gain from data. The iterative part of machine learning is significant in light of the fact that as models are presented to new information, they can independently adjust.

The original MNIST data is distributed through four binary files that you can download here. To demonstrate how nice Clojure is, I'm not using any special MNIST-specific code that is magically imported from the framework's model Zoo. The complete code, from scratch, is at the end of the article (I'm just pushing it there so it doesn't steal the spotlight:). The network learns in mini-batches of 128 images of the total of 60000, with adaptive moments, through 12 full epochs. That makes 5625 forward/backward update cycles.

GPT-3: Language Models are Few-Shot Learners May 29, 2020 Link Hacker News (431 points, 291 comments) Reddit (271 points, 113 comments) The original GPT-3 paper from OpenAI.

We present a framework that allows to certify the fairness degree of a model based on an interactive and privacy-preserving test. The framework verifies any trained model, regardless of its training process and architecture. Thus, it allows us to evaluate any deep learning model on multiple fairness definitions empirically. We tackle two scenarios, where either the test data is privately available only to the tester or is publicly known in advance, even to the model creator. We investigate the soundness of the proposed approach using theoretical analysis and present statistical guarantees for the interactive test.