As in many other scientific domains, we face a fundamental problem when using machine learning to identify proteins from mass spectrometry data: large ground truth datasets mapping inputs to correct outputs are extremely difficult to obtain. Instead, we have access to imperfect hand-coded models crafted by domain experts. In this paper, we apply deep neural networks to an important step of the protein identification problem, the pairing of mass spectra with short sequences of amino acids called peptides. We train our model to differentiate between top scoring results from a state-of-the art classical system and hard-negative second and third place results. Our resulting model is much better at identifying peptides with spectra than the model used to generate its training data. In particular, we achieve a 43% improvement over standard matching methods and a 10% improvement over a combination of the matching method and an industry standard cross-spectra reranking tool. Importantly, in a more difficult experimental regime that reflects current challenges facing biologists, our advantage over the previous state-of-the-art grows to 15% even after reranking. We believe this approach will generalize to other challenging scientific problems.

Capsule Networks have great potential to tackle problems in structural biology because of their attention to hierarchical relationships. This paper describes the implementation and application of a Capsule Network architecture to the classification of RAS protein family structures on GPU-based computational resources. The proposed Capsule Network trained on 2D and 3D structural encodings can successfully classify HRAS and KRAS structures. The Capsule Network can also classify a protein-based dataset derived from a PSI-BLAST search on sequences of KRAS and HRAS mutations. Our results show an accuracy improvement compared to traditional convolutional networks, while improving interpretability through visualization of activation vectors.

Topology optimization is computationally demanding that requires the assembly and solution to a finite element problem for each material distribution hypothesis. As a complementary alternative to the traditional physics-based topology optimization, we explore a data-driven approach that can quickly generate accurate solutions. To this end, we propose a deep learning approach based on a 3D encoder-decoder Convolutional Neural Network architecture for accelerating 3D topology optimization and to determine the optimal computational strategy for its deployment. Analysis of iteration-wise progress of the Solid Isotropic Material with Penalization process is used as a guideline to study how the earlier steps of the conventional topology optimization can be used as input for our approach to predict the final optimized output structure directly from this input. We conduct a comparative study between multiple strategies for training the neural network and assess the effect of using various input combinations for the CNN to finalize the strategy with the highest accuracy in predictions for practical deployment. For the best performing network, we achieved about 40% reduction in overall computation time while also attaining structural accuracies in the order of 96%.

Mini-batch stochastic gradient methods are the current state of the art for large-scale distributed training of neural networks and other machine learning models. However, they fail to adapt to a changing communication vs computation trade-off in a system, such as when scaling to a large number of workers or devices. More so, the fixed requirement of communication bandwidth for gradient exchange severely limits the scalability to multi-node training e.g. in datacenters, and even more so for training on decentralized networks such as mobile devices. We argue that variants of local SGD, which perform several update steps on a local model before communicating to other nodes, offer significantly improved overall performance and communication efficiency, as well as adaptivity to the underlying system resources. Furthermore, we present a new hierarchical extension of local SGD, and demonstrate that it can efficiently adapt to several levels of computation costs in a heterogeneous distributed system.

The exponential growth in the number of complex datasets every year requires more enhancement in machine learning methods to provide robust and accurate data classification. Lately, deep learning approaches have achieved surpassing results in comparison to previous machine learning algorithms. However, finding the suitable structure for these models has been a challenge for researchers. This paper introduces Random Multimodel Deep Learning (RMDL): a new ensemble, deep learning approach for classification. RMDL solves the problem of finding the best deep learning structure and architecture while simultaneously improving robustness and accuracy through ensembles of deep learning architectures. In short, RMDL trains multiple randomly generated models of Deep Neural Network (DNN), Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) in parallel and combines their results to produce better result of any of those models individually. In this paper, we describe RMDL model and compare the results for image and text classification as well as face recognition. We used MNIST and CIFAR-10 datasets as ground truth datasets for image classification and WOS, Reuters, IMDB, and 20newsgroup datasets for text classification. Lastly, we used ORL dataset to compare the model performance on face recognition task.

Deep Learning (DL) provides a methodology to predict extreme loads observed in energy grids. Forecasting energy loads and prices is challenging due to sharp peaks and troughs that arise from intraday system constraints due to supply and demand fluctuations. We propose deep spatio-temporal models and extreme value theory (DL-EVT) to capture the tail behavior of load spikes. Deep architectures, such as ReLU and LSTM can model generation trends and temporal dependencies while EVT captures highly volatile load spikes. To illustrate our methodology, we use hourly price and demand data from the PJM interconnection for 4719 nodes and we develop a deep predictor. DL-EVT outperforms traditional Fourier and time series methods, both in-and out-of-sample, by capturing the nonlinearities in prices. Finally, we conclude with directions for future research.

In this paper, we study how to make clustering benefiting from differentiable programming whose basic idea is treating the neural network as a language instead of a machine learning method. To this end, we recast the vanilla $k$-means as a novel feedforward neural network in an elegant way. Our contribution is two-fold. On the one hand, the proposed \textit{k}-meansNet is a neural network implementation of the vanilla \textit{k}-means, which enjoys four advantages highly desired, i.e., robustness to initialization, fast inference speed, the capability of handling new coming data, and provable convergence. On the other hand, this work may provide novel insights into differentiable programming. More specifically, most existing differentiable programming works unroll an \textbf{optimizer} as a \textbf{recurrent neural network}, namely, the neural network is employed to solve an existing optimization problem. In contrast, we reformulate the \textbf{objective function} of \textit{k}-means as a \textbf{feedforward neural network}, namely, we employ the neural network to describe a problem. In such a way, we advance the boundary of differentiable programming by treating the neural network as from an alternative optimization approach to the problem formulation. Extensive experimental studies show that our method achieves promising performance comparing with 12 clustering methods on some challenging datasets.

Many poker systems, whether created with heuristics or machine learning, rely on the probability of winning as a key input. However, calculating the precise probability using combinatorics is an intractable problem, so instead we approximate it. Monte Carlo simulation is an effective technique that can be used to approximate the probability that a player will win and/or tie a hand. However, without the use of a memory-intensive lookup table or a supercomputer, it becomes infeasible to run millions of times when training an agent with self-play. To combat the space-time tradeoff, we use deep learning to approximate the probabilities obtained from the Monte Carlo simulation with high accuracy. The learned model proves to be a lightweight alternative to Monte Carlo simulation, which ultimately allows us to use the probabilities as inputs during self-play efficiently.

Generative Adversarial Networks (GANs) have shown great capacity on image generation, in which a discriminative model guides the training of a generative model to construct images that resemble real images. Recently, GANs have been extended from generating images to generating sequences (e.g., poems, music and codes). Existing GANs on sequence generation mainly focus on general sequences, which are grammar-free. In many real-world applications, however, we need to generate sequences in a formal language with the constraint of its corresponding grammar. For example, to test the performance of a database, one may want to generate a collection of SQL queries, which are not only similar to the queries of real users, but also follow the SQL syntax of the target database. Generating such sequences is highly challenging because both the generator and discriminator of GANs need to consider the structure of the sequences and the given grammar in the formal language. To address these issues, we study the problem of syntax-aware sequence generation with GANs, in which a collection of real sequences and a set of pre-defined grammatical rules are given to both discriminator and generator. We propose a novel GAN framework, namely TreeGAN, to incorporate a given Context-Free Grammar (CFG) into the sequence generation process. In TreeGAN, the generator employs a recurrent neural network (RNN) to construct a parse tree. Each generated parse tree can then be translated to a valid sequence of the given grammar. The discriminator uses a tree-structured RNN to distinguish the generated trees from real trees. We show that TreeGAN can generate sequences for any CFG and its generation fully conforms with the given syntax. Experiments on synthetic and real data sets demonstrated that TreeGAN significantly improves the quality of the sequence generation in context-free languages.

Deep neural networks have recently achieved competitive accuracy for human activity recognition. However, there is room for improvement, especially in modeling long-term temporal importance and determining the activity relevance of different temporal segments in a video. To address this problem, we propose a learnable and differentiable module: Deep Adaptive Temporal Pooling (DATP). DATP applies a self-attention mechanism to adaptively pool the classification scores of different video segments. Specifically, using frame-level features, DATP regresses importance of different temporal segments and generates weights for them. Remarkably, DATP is trained using only the video-level label. There is no need of additional supervision except video-level activity class label. We conduct extensive experiments to investigate various input features and different weight models. Experimental results show that DATP can learn to assign large weights to key video segments. More importantly, DATP can improve training of frame-level feature extractor. This is because relevant temporal segments are assigned large weights during back-propagation. Overall, we achieve state-of-the-art performance on UCF101, HMDB51 and Kinetics datasets.