Complex problems may require sophisticated, non-linear learning methods such as kernel machines or deep neural networks to achieve state of the art prediction accuracies. However, high prediction accuracies are not the only objective to consider when solving problems using machine learning. Instead, particular scientific applications require some explanation of the learned prediction function. Unfortunately, most methods do not come with out of the box straight forward interpretation. Even linear prediction functions are not straight forward to explain if features exhibit complex correlation structure. In this paper, we propose the Measure of Feature Importance (MFI). MFI is general and can be applied to any arbitrary learning machine (including kernel machines and deep learning). MFI is intrinsically non-linear and can detect features that by itself are inconspicuous and only impact the prediction function through their interaction with other features. Lastly, MFI can be used for both --- model-based feature importance and instance-based feature importance (i.e, measuring the importance of a feature for a particular data point).

We present the Wright-Fisher Indian buffet process (WF-IBP), a probabilistic model for time-dependent data assumed to have been generated by an unknown number of latent features. This model is suitable as a prior in Bayesian nonparametric feature allocation models in which the features underlying the observed data exhibit a dependency structure over time. More specifically, we establish a new framework for generating dependent Indian buffet processes, where the Poisson random field model from population genetics is used as a way of constructing dependent beta processes. Inference in the model is complex, and we describe a sophisticated Markov Chain Monte Carlo algorithm for exact posterior simulation. We apply our construction to develop a nonparametric focused topic model for collections of time-stamped text documents and test it on the full corpus of NIPS papers published from 1987 to 2015.

In this work, we present an application of Locally Interpretable Machine-Agnostic Explanations to 2-D chemical structures. Using this framework we are able to provide a structural interpretation for an existing black-box model for classifying biologically produced fuel compounds with regard to Research Octane Number. This method of "painting" locally interpretable explanations onto 2-D chemical structures replicates the chemical intuition of synthetic chemists, allowing researchers in the field to directly accept, reject, inform and evaluate decisions underlying inscrutably complex quantitative structure-activity relationship models.

David Cheng is an investment manager at DCM Ventures where he focuses on opportunities in the consumer internet, applications and SaaS. In living organisms, evolution is a multi-generational process where mutations in genes are dropped and added. Well-adapted organisms survive and those less fortunate go extinct. Resilience is great, but if you don't grow gills in time for the flood, then tough luck. Engineering, on the other hand, is a deliberate process with reliable steps designed to reach a stated objective.

How could they see anything but the shadows if they were never allowed to move their heads? Python and high level libraries like Scikit-learn, TensorFlow, NLTK, PyBrain, Theano, and MLPY have made machine learning accessible to a broad programming community that might never have found it otherwise. With the democratization of these tools, there is now a large, and growing, population of machine learning practitioners who are primarily self-taught. At the same time, the stakes of machine learning have never been higher; predictive tools are driving decision-making in every sector, from business, art, and engineering to education, law, and defense. How do we ensure our predictions are valid and robust in a time when these few lines of Python can instantiate and fit a model?

It seems like AI, data science, machine learning and bots are some of the most discussed topics in tech today. My preferred way to do this is always through listening to podcasts. Here are the ones I've found the most interesting: They alternate between great interviews with academics & practitioners and short 10–15 minute episodes where the hosts give a short primer on topics like calculating feature importance, k-means clustering, natural language processing and decision trees, often using analogies related to their pet parrot, Yoshi. This is the only place where you'll learn about k-means clustering via placement of parrot droppings. Hosted by Katie Malone and Ben Jaffe, this weekly podcast covers diverse topics in data science and machine learning: talking about specific concepts like model theft and the cold start problem and how they apply to real-world problems and datasets.

The phase retrieval problem has garnered significant attention since the development of the PhaseLift algorithm, which is a convex program that operates in a lifted space of matrices. Because of the substantial computational cost due to lifting, many approaches to phase retrieval have been developed, including non-convex optimization algorithms which operate in the natural parameter space, such as Wirtinger Flow. Very recently, a convex formulation called PhaseMax has been discovered, and it has been proven to achieve phase retrieval via linear programming in the natural parameter space under optimal sample complexity. The current proofs of PhaseMax rely on statistical learning theory or geometric probability theory. Here, we present a short and elementary proof that PhaseMax exactly recovers real-valued vectors from random measurements under optimal sample complexity.

David Cheng is an investment manager at DCM Ventures where he focuses on opportunities in the consumer internet, applications and SaaS. In living organisms, evolution is a multi-generational process where mutations in genes are dropped and added. Well-adapted organisms survive and those less fortunate go extinct. Resilience is great, but if you don't grow gills in time for the flood, then tough luck. Engineering, on the other hand, is a deliberate process with reliable steps designed to reach a stated objective.

Survival analysis is a fundamental tool in medical research to identify predictors of adverse events and develop systems for clinical decision support. In order to leverage large amounts of patient data, efficient optimisation routines are paramount. We propose an efficient training algorithm for the kernel survival support vector machine (SSVM). We directly optimise the primal objective function and employ truncated Newton optimisation and order statistic trees to significantly lower computational costs compared to previous training algorithms, which require $O(n^4)$ space and $O(p n^6)$ time for datasets with $n$ samples and $p$ features. Our results demonstrate that our proposed optimisation scheme allows analysing data of a much larger scale with no loss in prediction performance. Experiments on synthetic and 5 real-world datasets show that our technique outperforms existing kernel SSVM formulations if the amount of right censoring is high ($\geq85\%$), and performs comparably otherwise.

Adaptive stochastic gradient methods such as AdaGrad have gained popularity in particular for training deep neural networks. The most commonly used and studied variant maintains a diagonal matrix approximation to second order information by accumulating past gradients which are used to tune the step size adaptively. In certain situations the full-matrix variant of AdaGrad is expected to attain better performance, however in high dimensions it is computationally impractical. We present Ada-LR and RadaGrad two computationally efficient approximations to full-matrix AdaGrad based on randomized dimensionality reduction. They are able to capture dependencies between features and achieve similar performance to full-matrix AdaGrad but at a much smaller computational cost. We show that the regret of Ada-LR is close to the regret of full-matrix AdaGrad which can have an up-to exponentially smaller dependence on the dimension than the diagonal variant. Empirically, we show that Ada-LR and RadaGrad perform similarly to full-matrix AdaGrad. On the task of training convolutional neural networks as well as recurrent neural networks, RadaGrad achieves faster convergence than diagonal AdaGrad.