Visual question answering (VQA) is a task that requires AI systems to display multi-modal understanding. A system must be able to reason over the question being asked as well as the image itself to determine reasonable answers to the questions posed. In many cases, simply reasoning over the image itself and the question is not enough to achieve good performance. As an aid of the task, other than region based visual information and natural language questions, external textual knowledge extracted from images can also be used to generate correct answers for questions. Considering these, we propose a deep neural network model that uses an attention mechanism which utilizes image features, the natural language question asked and semantic knowledge extracted from the image to produce open-ended answers for the given questions. The combination of image features and contextual information about the image bolster a model to more accurately respond to questions and potentially do so with less required training data. We evaluate our proposed architecture on a VQA task against a strong baseline and show that our method achieves excellent results on this task.

Cross entropy is the most widely used loss function for supervised training of image classification models. In this paper, we propose a novel training methodology that consistently outperforms cross entropy on supervised learning tasks across different architectures and data augmentations. We modify the batch contrastive loss, which has recently been shown to be very effective at learning powerful representations in the self-supervised setting. We are thus able to leverage label information more effectively than cross entropy. Clusters of points belonging to the same class are pulled together in embedding space, while simultaneously pushing apart clusters of samples from different classes. In addition to this, we leverage key ingredients such as large batch sizes and normalized embeddings, which have been shown to benefit self-supervised learning. On both ResNet-50 and ResNet-200, we outperform cross entropy by over 1%, setting a new state of the art number of 78.8% among methods that use AutoAugment data augmentation. The loss also shows clear benefits for robustness to natural corruptions on standard benchmarks on both calibration and accuracy. Compared to cross entropy, our supervised contrastive loss is more stable to hyperparameter settings such as optimizers or data augmentations.

We train an artificial neural network which distinguishes chaotic and regular dynamics of the two-dimensional Chirikov standard map. We use finite length trajectories and compare the performance with traditional numerical methods which need to evaluate the Lyapunov exponent. The neural network has superior performance for short periods with length down to 10 Lyapunov times on which the traditional Lyapunov exponent computation is far from converging. We show the robustness of the neural network to varying control parameters, in particular we train with one set of control parameters, and successfully test in a complementary set. Furthermore, we use the neural network to successfully test the dynamics of discrete maps in different dimensions, e.g. the one-dimensional logistic map and a three-dimensional discrete version of the Lorenz system. Our results demonstrate that a convolutional neural network can be used as an excellent chaos indicator.

Self-supervised learning is one of those recent ML methods that have caused a ripple effect in the data science network, yet have so far been flying under the radar to the extent Entrepreneurs and Fortunes of the world go; the overall population is yet to find out about the idea yet lots of AI society consider it progressive. The paradigm holds immense potential for enterprises too as it can help handle deep learning's most overwhelming issue: data/sample inefficiency and subsequent costly training. Yann LeCun said that if knowledge was a cake, unsupervised learning would be the cake, supervised learning would be the icing on the cake and reinforcement learning would be the cherry on the cake. We realize how to make the icing and the cherry, however, we don't have a clue how to make the cake." Unsupervised learning won't progress a lot and said there is by all accounts a massive conceptual disconnect with regards to how precisely it should function and that it was the dark issue of ...

Any data scientist or machine learning enthusiast would have heard, at least once in their life, that Deep Learning requires a lot of hardware. Some train simple deep learning models for days on their laptops (typically without GPUs) which leads to an impression that Deep Learning requires big systems to run execute. This has created a myth surrounding deep learning which creates a roadblock for beginners. Every book that I've referred to in the past few years has the author always mentioning the following: Deep learning requires a lot of computational power to run on. But I don't have datacentres at my command and when I built my first deep learning model on a sizable laptop, I knew that the consensus was either wrong or portrayed with some truth.

Which is the best framework for programming Deep Learning networks?

Since the earliest days of artificial intelligence -- and computing more generally -- theorists have assumed that intelligent machines would think in much the same ways as humans. After all, we know of no greater cognitive power than the human brain. In many ways, it makes sense to try to replicate it if the goal is to create a high level of cognitive processing. However, there is a debate today over the best way of reaching true general AI. In particular, recent years' advancements in deep learning -- which is itself inspired by the human brain, though diverges from it in some important ways -- have shown developers that there may be other paths.

Autonomous detection of desired events from large databases using time series classification is becoming increasingly important in civil engineering as a result of continued long-term health monitoring of a large number of engineering structures encompassing buildings, bridges, towers, and offshore platforms. In this context, this paper proposes the application of a relatively new time series representation named "Shapelet transform", which is based on local similarity in the shape of the time series subsequences. In consideration of the individual attributes distinctive to time series signals in earthquake, wind and ocean engineering, the application of this transform yields a new shape-based feature representation. Combining this shape-based representation with a standard machine learning algorithm, a truly "white-box" machine learning model is proposed with understandable features and a transparent algorithm. This model automates event detection without the intervention of domain practitioners, yielding a practical event detection procedure. The efficacy of this proposed shapelet transform-based autonomous detection procedure is demonstrated by examples, to identify known and unknown earthquake events from continuously recorded ground-motion measurements, to detect pulses in the velocity time history of ground motions to distinguish between near-field and far-field ground motions, to identify thunderstorms from continuous wind speed measurements, to detect large-amplitude wind-induced vibrations from the bridge monitoring data, and to identify plunging breaking waves that have a significant impact on offshore structures.

This work presents a Long Short-Term Memory (LSTM) network for forecasting a monthly electricity demand time series with a one-year horizon. The novelty of this work is the use of pattern representation of the seasonal time series as an alternative to decomposition. Pattern representation simplifies the complex nonlinear and nonstationary time series, filtering out the trend and equalizing variance. Two types of patterns are defined: x-pattern and y-pattern. The former requires additional forecasting for the coding variables. The latter determines the coding variables from the process history. A hybrid approach based on x-patterns turned out to be more accurate than the standard LSTM approach based on a raw time series. In this combined approach an x-pattern is forecasted using a sequence-to-sequence LSTM network and the coding variables are forecasted using exponential smoothing. A simulation study performed on the monthly electricity demand time series for 35 European countries confirmed the high performance of the proposed model and its competitiveness to classical models such as ARIMA and exponential smoothing as well as the MLP neural network model.

Creating a state-of-the-art deep-learning system requires vast amounts of data, expertise, and hardware, yet research into embedding copyright protection for neural networks has been limited. One of the main methods for achieving such protection involves relying on the susceptibility of neural networks to backdoor attacks, but the robustness of these tactics has been primarily evaluated against pruning, fine-tuning, and model inversion attacks. In this work, we propose a neural network "laundering" algorithm to remove black-box backdoor watermarks from neural networks even when the adversary has no prior knowledge of the structure of the watermark. We are able to effectively remove watermarks used for recent defense or copyright protection mechanisms while achieving test accuracies above 97% and 80% for both MNIST and CIFAR-10, respectively. For all backdoor watermarking methods addressed in this paper, we find that the robustness of the watermark is significantly weaker than the original claims. We also demonstrate the feasibility of our algorithm in more complex tasks as well as in more realistic scenarios where the adversary is able to carry out efficient laundering attacks using less than 1% of the original training set size, demonstrating that existing backdoor watermarks are not sufficient to reach their claims.