Michael C. Mozer

Bayesian community-wide culture-independent microbial source tracking

Contamination is a critical issue in high-throughput metagenomic studies, yet progress toward a comprehensive solution has been limited. We present SourceTracker, a Bayesian approach to estimate the proportion of contaminants in a given community that come from possible source environments. We applied SourceTracker to microbial surveys from neonatal intensive care units (NICUs), offices and molecular biology laboratories, and provide a database of known contaminants for future testing.

Predicting subscriber dissatisfaction and improving retention in the wireless telecommunications industry

Competition in the wireless telecommunications industry is fierce. To maintain profitability, wireless carriers must control churn, which is the loss of subscribers who switch from one carrier to another.We explore techniques from statistical machine learning to predict churn and, based on these predictions, to determine what incentives should be offered to subscribers to improve retention and maximize profitability to the carrier. The techniques include logit regression, decision trees, neural networks, and boosting. Our experiments are based on a database of nearly 47,000 U.S. domestic subscribers and includes information about their usage, billing, credit, application, and complaint history. Our experiments show that under a wide variety of assumptions concerning the cost of intervention and the retention rate resulting from intervention, using predictive techniques to identify potential churners and offering incentives can yield significant savings to a carrier. We also show the importance of a data representation crafted by domain experts. Finally, we report on a real-world test of the techniques that validate our simulation experiments.

Using Relevance to Reduce Network Size Automatically

This paper proposes a means of using the knowledge in a network to determine the functionality or relevance of individual units, both for the purpose of understanding the networks behavior and imp...

Object-Based Attention and Occlusion Evidence From Normal Participants and a Computational Model

One way of perceptually organizing a complex visual scene is to attend selectively to information in a particular physical location. Another way of reducing the complexity in the input is to attend selectively to an individual object in the scene and to process its elements preferentially. This latter, object-based attention process was examined, and the predicted superiority for reporting features from 1 relative to 2 objects was replicated in a series of experiments. This object-based process was robust even under conditions of occlusion, although there were some boundary conditions on its operation. Finally, an account of the data is provided via simulations of the findings in a computational model. The claim is that object-based attention arises from a mechanisms that groups together those features based on internal representations developed over perceptual experience and then preferentially gates these features for later, selective processing.

An Intelligent Environment Must Be Adaptive

Michael C. Mozer, University of Colorado What will the home of the future look like? One popular vision is that household devices-appliances, entertainment centers, phones, thermostats, lights-will be endowed with microprocessors that allow the devices to communicate with one another and with the home’s inhabitants. The dishwasher can ask the water heater whether the water temperature is adequate; inhabitants can telephone home and remotely instruct the VCR to record a favonte show; the TV could select news stones of special interest to the inhabitant; the stereo might lower its volume when the phone rings; and the clothes dryer might make an announcement over an intercom system when it has completed its cycle.

On the interaction of selective attention and lexical knowledge: A connectionist account of neglect dyslexia

Neglect dyslexia, a reading impairment acquired as a consequence of brain injury, is traditionally interpreted as a disturbance of selective attention. Patients with neglect dyslexia may ignore the left side of an open book, the beginning words of a line of text, or the beginning letters of a single word. These patients provide a rich but sometimes contradictory source of data regarding the locus of attentional selectivity. We have reconsidered the patient data within the framework of an existing connectionist model of word recognition and spatial attention. We show that the effects of damage to the model resemble the reading impairments observed in neglect dyslexia. In simulation experiments, we account for a broad spectrum of behaviors including the following: (1) when two noncontiguous stimuli are presented simultaneously, the contralesional stimulus is neglected (extinction); (2) explicit instructions to the patient can reduce the severity of neglect; (3) stimulus position in the visual field affects reading performance; (4) words are read much better than pronounceable nonwords; (5) the nature of error responses depends on the morphemic composition of the stimulus; and (6) extinction interacts with lexical knowledge (if two words are presented that form a compound, e.g., COW and BOY, the patient is more likely to report both than in a control condition, e.g., SUN and FLY). The convergence of findings from the neuropsychological research and the computational modeling sheds light on the role of attention in normal visuospatial processing, supporting a hybrid view of attentional selection that has properties of both early and late selection.

Letter Migration in Word Perception.

These experiments demonstrate that the perception of two distinct words in a briefLy presented display can interact, causing perceptual migrations of letters from one word to the other. For example, when LINE and LACE are presented, subjects might report seeing LICE or LANE instead of LINE. Several properties of the letter migrations were revealed: (a) Migrations are more frequent when the words are separated by smaller physical distances; (b) a majority of the migrations are a result of letters being copied from one word to the other, not from the interchange of letters of the two words; (c) migrations to a word are less frequent when subjects focus attention on that word; and (d) migrations are far more frequent when the words share letters in common. This last result suggests that migrations are not caused by a loss of spatial information at the letter level, that is, by free-floating letters being wrongly combined. Rather, migrations occur because of structural limitations at a high level of the word-recognition process, perhaps during lexical activation. Implications for models of multiple-word perception are discussed.

Optimizing Distributed Practice Theoretical Analysis and Practical Implications

More than a century of research shows that increasing the gap between study episodes using the same material can enhance retention, yet little is known about how this so-called distributed practice effect unfolds over nontrivial periods. In two three-session laboratory studies, we examined the effects of gap on retention of foreign vocabulary, facts, and names of visual objects, with test delays up to 6 months. An optimal gap improved final recall by up to 150%. Both studies demonstrated nonmonotonic gap effects: Increases in gap caused test accuracy to initially sharply increase and then gradually decline. These results provide new constraints on theories of spacing and confirm the importance of cumulative reviews to promote retention over meaningful time periods.

Neural network music composition by prediction: exploring the benefits of psychoacoustic constraints and multi-scale processing

Abstract In algorithmic music composition, a simple technique involves selecting notes sequentially according to a transition table that specifies the probability of the next note as a function of the previous context. An extension of this transition-table approach is described, using a recurrent autopredictive connectionist network called CONCERT. CONCERT is trained on a set of pieces with the aim of extracting stylistic regularities. CONCERT can then be used to compose new pieces. A central ingredient of CONCERT is the incorporation of psychologically grounded representations of pitch, duration and harmonic structure. CONCERT was tested on sets of examples artificially generated according to simple rules and was shown to learn the underlying structure, even where other approaches failed. In larger experiments, CONCERT was trained on sets of J. S. Bach pieces and traditional European folk melodies and was then allowed to compose novel melodies. Although the compositions are occasionally pleasant, and are...

Evidence-based static branch prediction using machine learning

Correctly predicting the direction that branches will take is increasingly important in todays wide-issue computer architectures. The name program-based branch prediction is given to static branch prediction techniques that base their prediction on a programs structure. In this article, we investigate a new approach to program-based branch prediction that uses a body of existing programs to predict the branch behavior in a new program. We call this approach to program-based branch prediction evidence-based static prediction, or ESP. The main idea of ESP is that the behavior of a corpus of programs can be used to infer the behavior of new programs. In this article, we use neural networks and decision trees to map static features associated with each branch to a prediction that the branch will be taken. ESP shows significant advantages over other prediction mechanisms. Specifically, it is a program-based technique; it is effective across a range of programming languages and programming styles; and it does not rely on the use of expert-defined heuristics. In this article, we describe the application of ESP to the problem of static branch prediction and compare our results to existing program-based branch predictors. We also investigate the applicability of ESP across computer architectures, programming languages, compilers, and run-time systems. We provide results showing how sensitive ESP is to the number and type of static features and programs included in the ESP training sets, and we compare the efficacy of static branch prediction for subroutine libraries. Averaging over a body of 43 C and Fortran programs, ESP branch prediction results in a miss rate of 20%, as compared with the 25% miss rate obtained using the best existing program-based heuristics.