--Gene annotation has traditionally required direct comparison of DNA sequences between an unknown gene and a database of known ones using string comparison methods. However, these methods do not provide useful information when a gene does not have a close match in the database. In addition, each comparison can be costly when the database is large since it requires alignments and a series of string comparisons. In this work we propose a novel approach: using recurrent neural networks to embed DNA or amino-acid sequences in a low-dimensional space in which distances correlate with functional similarity. This embedding space overcomes both shortcomings of the method of aligning sequences and comparing homology. First, it allows us to obtain information about genes which do not have exact matches by measuring their similarity to other ones in the database. If our database is labeled this can provide labels for a query gene as is done in traditional methods. However, even if the database is unlabeled it allows us to find clusters and infer some characteristics of the gene population. In addition, each comparison is much faster than traditional methods since the distance metric is reduced to the Euclidean distance, and thus efficient approximate nearest neighbor algorithms can be used to find the best match. More specifically we show how our embedding can be useful for both classification tasks when our labels are known, and clustering tasks where our sequences belong to classes which have not been seen before. The central dogma of biology states that all organisms contain DNA, which is transcribed into RNA and then translated into proteins, which catalyze the chemical reactions that define life.

Choosing appropriate step sizes is critical for reducing the computational cost of training large-scale neural network models. Mini-batch sub-sampling (MBSS) is often employed for computational tractability. However, MBSS introduces a sampling error, that can manifest as a bias or variance in a line search. This is because MBSS can be performed statically, where the mini-batch is updated only when the search direction changes, or dynamically, where the mini-batch is updated every-time the function is evaluated. Static MBSS results in a smooth loss function along a search direction, reflecting low variance but large bias in the estimated "true" (or full batch) minimum. Conversely, dynamic MBSS results in a point-wise discontinuous function, with computable gradients using backpropagation, along a search direction, reflecting high variance but lower bias in the estimated "true" (or full batch) minimum. In this study, quadratic line search approximations are considered to study the quality of function and derivative information to construct approximations for dynamic MBSS loss functions. An empirical study is conducted where function and derivative information are enforced in various ways for the quadratic approximations. The results for various neural network problems show that being selective on what information is enforced helps to reduce the variance of predicted step sizes.

Deep neural networks (DNNs) are notorious for their vulnerability to adversarial attacks, which are small perturbations added to their input images to mislead their prediction. Detection of adversarial examples is, therefore, a fundamental requirement for robust classification frameworks. In this work, we present a method for detecting such adversarial attacks, which is suitable for any pre-trained neural network classifier. We use influence functions to measure the impact of every training sample on the validation set data. From the influence scores, we find the most supportive training samples for any given validation example. A k-nearest neighbor (k-NN) model fitted on the DNN's activation layers is employed to search for the ranking of these supporting training samples. We observe that these samples are highly correlated with the nearest neighbors of the normal inputs, while this correlation is much weaker for adversarial inputs. We train an adversarial detector using the k-NN ranks and distances and show that it successfully distinguishes adversarial examples, getting state-of-the-art results on four attack methods with three datasets.

User churn is an important issue in online services that threatens the health and profitability of services. Most of the previous works on churn prediction convert the problem into a binary classification task where the users are labeled as churned and non-churned. More recently, some works have tried to convert the user churn prediction problem into the prediction of user return time. In this approach which is more realistic in real world online services, at each time-step the model predicts the user return time instead of predicting a churn label. However, the previous works in this category suffer from lack of generality and require high computational complexity. In this paper, we introduce \emph{ChOracle}, an oracle that predicts the user churn by modeling the user return times to service by utilizing a combination of Temporal Point Processes and Recurrent Neural Networks. Moreover, we incorporate latent variables into the proposed recurrent neural network to model the latent user loyalty to the system. We also develop an efficient approximate variational algorithm for learning parameters of the proposed RNN by using back propagation through time. Finally, we demonstrate the superior performance of ChOracle on a wide variety of real world datasets.

When estimating the relevancy between a query and a document, ranking models largely neglect the mutual information among documents. A common wisdom is that if two documents are similar in terms of the same query, they are more likely to have similar relevance score. To mitigate this problem, in this paper, we propose a multi-agent reinforced ranking model, named MarlRank. In particular, by considering each document as an agent, we formulate the ranking process as a multi-agent Markov Decision Process (MDP), where the mutual interactions among documents are incorporated in the ranking process. To compute the ranking list, each document predicts its relevance to a query considering not only its own query-document features but also its similar documents features and actions. By defining reward as a function of NDCG, we can optimize our model directly on the ranking performance measure. Our experimental results on two LETOR benchmark datasets show that our model has significant performance gains over the state-of-art baselines. We also find that the NDCG shows an overall increasing trend along with the step of interactions, which demonstrates that the mutual information among documents helps improve the ranking performance.

Reinforcement learning frameworks have introduced abstractions to implement and execute algorithms at scale. They assume standardized simulator interfaces but are not concerned with identifying suitable task representations. We present Wield, a first-of-its kind system to facilitate task design for practical reinforcement learning. Through software primitives, Wield enables practitioners to decouple system-interface and deployment-specific configuration from state and action design. To guide experimentation, Wield further introduces a novel task design protocol and classification scheme centred around staged randomization to incrementally evaluate model capabilities.

The goal of imitation learning (IL) is to learn a good policy from high-quality demonstrations. However, the quality of demonstrations in reality can be diverse, since it is easier and cheaper to collect demonstrations from a mix of experts and amateurs. IL in such situations can be challenging, especially when the level of demonstrators' expertise is unknown. We propose a new IL method called \underline{v}ariational \underline{i}mitation \underline{l}earning with \underline{d}iverse-quality demonstrations (VILD), where we explicitly model the level of demonstrators' expertise with a probabilistic graphical model and estimate it along with a reward function. We show that a naive approach to estimation is not suitable to large state and action spaces, and fix its issues by using a variational approach which can be easily implemented using existing reinforcement learning methods. Experiments on continuous-control benchmarks demonstrate that VILD outperforms state-of-the-art methods. Our work enables scalable and data-efficient IL under more realistic settings than before.

We present a new explainable AI (XAI) framework aimed at increasing justified human trust and reliance in the AI machine through explanations. We pose explanation as an iterative communication process, i.e. dialog, between the machine and human user. More concretely, the machine generates sequence of explanations in a dialog which takes into account three important aspects at each dialog turn: (a) human's intention (or curiosity); (b) human's understanding of the machine; and (c) machine's understanding of the human user. To do this, we use Theory of Mind (ToM) which helps us in explicitly modeling human's intention, machine's mind as inferred by the human as well as human's mind as inferred by the machine. In other words, these explicit mental representations in ToM are incorporated to learn an optimal explanation policy that takes into account human's perception and beliefs. Furthermore, we also show that ToM facilitates in quantitatively measuring justified human trust in the machine by comparing all the three mental representations. We applied our framework to three visual recognition tasks, namely, image classification, action recognition, and human body pose estimation. We argue that our ToM based explanations are practical and more natural for both expert and non-expert users to understand the internal workings of complex machine learning models. To the best of our knowledge, this is the first work to derive explanations using ToM. Extensive human study experiments verify our hypotheses, showing that the proposed explanations significantly outperform the state-of-the-art XAI methods in terms of all the standard quantitative and qualitative XAI evaluation metrics including human trust, reliance, and explanation satisfaction.

With the increased sophistication of AI techniques, the application of these systems has been expanding to ever newer fields. Increasingly, these systems are being used in modeling of human aesthetics and creativity, e.g. how humans create artworks and design products. Our lab has developed one such AI creativity deep learning system that can be used to create artworks in the form of images and videos. In this paper, we describe this system and its use in studying the human visual system and the formation of aesthetic experiences. Specifically, we show how time-based AI created media can be used to explore the nature of the dual-pathway neuro-architecture of the human visual system and how this relates to higher cognitive judgments such as aesthetic experiences that rely on these divergent information streams. We propose a theoretical framework for how the movement within percepts such as video clips, causes the engagement of reflexive attention and a subsequent focus on visual information that are primarily processed via the dorsal stream, thereby modulating aesthetic experiences that rely on information relayed via the ventral stream. We outline our recent study in support of our proposed framework, which serves as the first study that investigates the relationship between the two visual streams and aesthetic experiences.

Could you imagine a future where computers made economic decisions rather than governments and central bankers? With all of the economic mishaps we've been seeing over the past decade, one could say it isn't a particularly bad idea! Natural language processing could allow us to make more sense of the economy than we do currently. As it stands, investors and policymakers use index benchmarks and quantitative measures such as GDP growth to gauge economic health. That said, one potential application of NLP is to analyse text data (such as through major economic policy documents), and then "learn" from such texts in order to generate appropriate economic policies independently of human intervention. In this example, an LSTM model is trained using text from a sample ECB policy document, in order to generate "new" text data, with a view to revealing insights from such text that could be used for policy purposes.