This paper describes and motivates a new decision theory known as functional decision theory (FDT), as distinct from causal decision theory and evidential decision theory. Functional decision theorists hold that the normative principle for action is to treat one's decision as the output of a fixed mathematical function that answers the question, "Which output of this very function would yield the best outcome?" Adhering to this principle delivers a number of benefits, including the ability to maximize wealth in an array of traditional decision-theoretic and game-theoretic problems where CDT and EDT perform poorly. Using one simple and coherent decision rule, functional decision theorists (for example) achieve more utility than CDT on Newcomb's problem, more utility than EDT on the smoking lesion problem, and more utility than both in Parfit's hitchhiker problem. In this paper, we define FDT, explore its prescriptions in a number of different decision problems, compare it to CDT and EDT, and give philosophical justifications for FDT as a normative theory of decision-making.

Super-resolution fluorescence microscopy, with a resolution beyond the diffraction limit of light, has become an indispensable tool to directly visualize biological structures in living cells at a nanometer-scale resolution. Despite advances in high-density super-resolution fluorescent techniques, existing methods still have bottlenecks, including extremely long execution time, artificial thinning and thickening of structures, and lack of ability to capture latent structures. Here we propose a novel deep learning guided Bayesian inference approach, DLBI, for the time-series analysis of high-density fluorescent images. Our method combines the strength of deep learning and statistical inference, where deep learning captures the underlying distribution of the fluorophores that are consistent with the observed time-series fluorescent images by exploring local features and correlation along time-axis, and statistical inference further refines the ultrastructure extracted by deep learning and endues physical meaning to the final image. Comprehensive experimental results on both real and simulated datasets demonstrate that our method provides more accurate and realistic local patch and large-field reconstruction than the state-of-the-art method, the 3B analysis, while our method is more than two orders of magnitude faster. The main program is available at https://github.com/lykaust15/DLBI

This paper introduces a novel Hilbert space representation of a counterfactual distribution---called counterfactual mean embedding (CME)---with applications in nonparametric causal inference. Counterfactual prediction has become an ubiquitous tool in machine learning applications, such as online advertisement, recommendation systems, and medical diagnosis, whose performance relies on certain interventions. To infer the outcomes of such interventions, we propose to embed the associated counterfactual distribution into a reproducing kernel Hilbert space (RKHS) endowed with a positive definite kernel. Under appropriate assumptions, the CME allows us to perform causal inference over the entire landscape of the counterfactual distribution. The CME can be estimated consistently from observational data without requiring any parametric assumption about the underlying distributions. We also derive a rate of convergence which depends on the smoothness of the conditional mean and the Radon-Nikodym derivative of the underlying marginal distributions. Our framework can deal with not only real-valued outcome, but potentially also more complex and structured outcomes such as images, sequences, and graphs. Lastly, our experimental results on off-policy evaluation tasks demonstrate the advantages of the proposed estimator.

Unsupervised learning is widely recognized as one of the most important challenges facing machine learning nowa- days. However, in spite of hundreds of papers on the topic being published every year, current theoretical understanding and practical implementations of such tasks, in particular of clustering, is very rudimentary. This note focuses on clustering. I claim that the most signif- icant challenge for clustering is model selection. In contrast with other common computational tasks, for clustering, dif- ferent algorithms often yield drastically different outcomes. Therefore, the choice of a clustering algorithm, and their pa- rameters (like the number of clusters) may play a crucial role in the usefulness of an output clustering solution. However, currently there exists no methodical guidance for clustering tool-selection for a given clustering task. Practitioners pick the algorithms they use without awareness to the implications of their choices and the vast majority of theory of clustering papers focus on providing savings to the resources needed to solve optimization problems that arise from picking some concrete clustering objective. Saving that pale in com- parison to the costs of mismatch between those objectives and the intended use of clustering results. I argue the severity of this problem and describe some recent proposals aiming to address this crucial lacuna.

We present a statistical mechanics model of deep feed forward neural networks (FFN). Our energy-based approach naturally explains several known results and heuristics, providing a solid theoretical framework and new instruments for a systematic development of FFN. We infer that FFN can be understood as performing three basic steps: encoding, representation validation and propagation. We obtain a set of natural activations - such as sigmoid, tanh and ReLu - together with a state-of-the-art one, recently obtained by Ramachandran et al. [1] using an extensive search algorithm. We term this activation ESP (Expected Signal Propagation), explain its probabilistic meaning, and study the eigenvalue spectrum of the associated Hessian on classification tasks. We find that ESP allows for faster training and more consistent performances over a wide range of network architectures.

In this paper, we propose a cost-aware cascading bandits model, a new variant of multi-armed ban- dits with cascading feedback, by considering the random cost of pulling arms. In each step, the learning agent chooses an ordered list of items and examines them sequentially, until certain stopping condition is satisfied. Our objective is then to max- imize the expected net reward in each step, i.e., the reward obtained in each step minus the total cost in- curred in examining the items, by deciding the or- dered list of items, as well as when to stop examina- tion. We study both the offline and online settings, depending on whether the state and cost statistics of the items are known beforehand. For the of- fline setting, we show that the Unit Cost Ranking with Threshold 1 (UCR-T1) policy is optimal. For the online setting, we propose a Cost-aware Cas- cading Upper Confidence Bound (CC-UCB) algo- rithm, and show that the cumulative regret scales in O(log T ). We also provide a lower bound for all {\alpha}-consistent policies, which scales in {\Omega}(log T ) and matches our upper bound. The performance of the CC-UCB algorithm is evaluated with both synthetic and real-world data.

Image has been cropped and resized. Scientists have developed machine-learning algorithms that can identify patterns in the bacteria of a patient's gut to determine whether the patient is likely to get infected if exposed to cholera. The researchers believe such artificial intelligence (AI) could be critical in areas of high cholera risk, since it can analyze trillions of bacteria, much more than can be done by humans. The study also demonstrates the power of machine learning to uncover medical insights that would otherwise remain obscure. READ: AI's Ethical Concerns Go Beyond Data Security and Quality The research is a collaboration between Duke University, Massachusetts General Hospital, and the International Centre for Diarrheal Disease Research, in Bangladesh.

These neuropsychiatric disorders are prevalent in low- to middle-income countries due to various factors, e.g. Around 80 percent of the world's epilepsy occurs in low- to middle-income countries, but only 20 percent of people get treatment. The physician-to-patient ratio can be as low as one for every 20,000 people in those countries, with even fewer psychiatrists and neurologists, causing a so-called treatment gap.1 However, timely diagnosis and treatment of epilepsy is possible and can make a difference.2 Last fall, partnering with the Nanyang Technological University (NTU) of Singapore, we took the first steps in tackling this challenge in our Science for Social Good program. Our team included a Social Good Fellow from Columbia University, several machine learning and cloud computing researchers from IBM Research, and collaborators from NTU. Together, we came up with a cloud-based automated machine learning approach to provide decision support for non-specialist physicians in electroencephalography (EEG) analysis and interpretation.

Baidu's billionaire founder Robin Li Yanhong held a town hall in Beijing on Monday in which he told those present that he remained confident in the company's prospects and that the strategy to transform itself into an artificial intelligence (AI) company remains unchanged with the impending departure of chief operating officer Lu Qi, according to people with knowledge of the meeting. Lu was present at the meeting and said he is leaving the company for personal and family reasons, said the people, who asked not to be identified discussing the company's internal matters. Baidu declined to verify what was said at the meeting. Li, chairman and chief executive of China's dominant search engine operator, sought to address concerns after news on Friday that Lu would be stepping down, in another high-profile departure for Baidu, which is seeking to transform itself into a leader in AI. On Friday, the Nasdaq-listed company shocked China's tech universe by announcing the departure of Lu, a former Microsoft executive who was hired about 18 months ago to steer the company's transformation into an AI-driven powerhouse.

It wasn't too long ago that the concept of carrying a sophisticated computer, camera and phone, all rolled into one gadget fitting in your pocket, was the stuff of science fiction. Now smartphones are everywhere and they're getting smarter all the time. Imagine when your phone will be able to diagnose most of your medical problems for you based on artificial intelligence (AI) in the cloud, saving you a trip to the doctor. The app could issue a diagnosis and a prescription, and your local pharmacy could 3D-print your medicine. This exciting new frontier is part of the Fourth Industrial Revolution (4IR), a period of rapid change driven by progress in science and technology.