Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Functional neuroimaging studies have started unravelling unexpected functional attributes for the posteromedial portion of the parietal lobe, the precuneus. This cortical area has traditionally received little attention, mainly because of its hidden location and the virtual absence of focal lesion studies. However, recent functional imaging findings in healthy subjects suggest a central role for the precuneus in a wide spectrum of highly integrated tasks, including visuo-spatial imagery, episodic memory retrieval and self-processing operations, namely first-person perspective taking and an experience of agency. Furthermore, precuneus and surrounding posteromedial areas are amongst the brain structures displaying the highest resting metabolic rates (hot spots) and are characterized by transient decreases in the tonic activity during engagement in non-self-referential goal-directed actions (default mode of brain function). Therefore, it has recently been proposed that precuneus is involved in the interwoven network of the neural correlates of self-consciousness, engaged in self-related mental representations during rest. This hypothesis is consistent with the selective hypometabolism in the posteromedial cortex reported in a wide range of altered conscious states, such as sleep, drug-induced anaesthesia and vegetative states. This review summarizes the current knowledge about the macroscopic and microscopic anatomy of precuneus, together with its wide-spread connectivity with both cortical and subcortical structures, as shown by connectional and neurophysiological findings in non-human primates, and links these notions with the multifaceted spectrum of its behavioural correlates. By means of a critical analysis of precuneus activation patterns in response to different mental tasks, this paper provides a useful conceptual framework for matching the functional imaging findings with the specific role(s) played by this structure in the higher-order cognitive functions in which it has been implicated. Specifically, activation patterns appear to converge with anatomical and connectivity data in providing preliminary evidence for a functional subdivision within the precuneus into an anterior region, involved in self-centred mental imagery strategies, and a posterior region, subserving successful episodic memory retrieval.

/pdf/the-precuneus-a-review-of-its-functional-anatomy-and-q9xsn3nfnb.pdf

The precuneus: a review of its functional anatomy and behavioural correlates.

Schultz, Wolfram. Predictive reward signal of dopamine neurons. J. Neurophysiol. 80: 1–27, 1998. The effects of lesions, receptor blocking, electrical self-stimulation, and drugs of abuse suggest t...

/pdf/predictive-reward-signal-of-dopamine-neurons-3f94myirza.pdf

Predictive Reward Signal of Dopamine Neurons

The connection matrix of the human brain (the human “connectome”) represents an indispensable foundation for basic and applied neurobiological research. However, the network of anatomical connections linking the neuronal elements of the human brain is still largely unknown. While some databases or collations of large-scale anatomical connection patterns exist for other mammalian species, there is currently no connection matrix of the human brain, nor is there a coordinated research effort to collect, archive, and disseminate this important information. We propose a research strategy to achieve this goal, and discuss its potential impact.

/pdf/the-human-connectome-a-structural-description-of-the-human-31vln7dys2.pdf

The Human Connectome: A Structural Description of the Human Brain

A key problem in robot grasping is that of positioning the manipulator contacts so that an object can be grasped. In unstructured environments, contact positions are typically planned based on range or visual measurements that are used to reconstruct object geometry. However, because it is difficult to measure the complete object geometry precisely in common grasp scenarios, it is useful to employ additional techniques to adjust or refine the grasp using local information only. In particular, grasp control techniques can be used to improve a grasp by adjusting the contact configuration after making initial contact with an object by using measurements of local object geometry at the contacts. This paper proposes three variations on null-space grasp control: an approach that combines multiple grasp objectives to improve a grasp. Two of these variations are theoretically demonstrated to converge to force-closure configurations for arbitrary convex objects when grasping with two contacts. All variations are found to converge in simulation. Robot-grasping experiments are reported that show the approach to be useful in practice.

/pdf/null-space-grasp-control-theory-and-experiments-sfx8jk3r75.pdf

Null-Space Grasp Control: Theory and Experiments

In dexterous manipulation, an object must be reconfigured while maintaining a stable grasp. This may require that the object be re-grasped in order to avoid finger workspace limits. We present a set of closed-loop controllers designed to achieve force-related objectives such as wrench closure, and show how they may be concurrently combined. Furthermore, we show that dexterous manipulation behavior may be generated by sequencing concurrent combinations of these controllers. We show that dexterous manipulation can be viewed as a task that is accomplished in the context of a wrench closure constraint. We hypothesize this approach can generalize to any task that must be accomplished while maintaining a set of constraints.

Manipulation gaits: sequences of grasp control tasks

Humanoid robots promise to lead us toward more effective and informative interactions between humans and robotic devices. The authors introduce mechanisms for attention control and pattern categorization as the basis for cognition in a humanoid robot.

/pdf/tracing-patterns-and-attention-humanoid-robot-cognition-52hbz9dlx4.pdf

Tracing patterns and attention: humanoid robot cognition

Objective. A brain–machine interface (BMI) records neural signals in real time from a subject's brain, interprets them as motor commands, and reroutes them to a device such as a robotic arm, so as to restore lost motor function. Our objective here is to improve BMI performance by minimizing the deleterious effects of delay in the BMI control loop. We mitigate the effects of delay by decoding the subject's intended movements a short time lead in the future. Approach. We use the decoded, intended future movements of the subject as the control signal that drives the movement of our BMI. This should allow the user's intended trajectory to be implemented more quickly by the BMI, reducing the amount of delay in the system. In our experiment, a monkey (Macaca mulatta) uses a future prediction BMI to control a simulated arm to hit targets on a screen. Main Results. Results from experiments with BMIs possessing different system delays (100, 200 and 300 ms) show that the monkey can make significantly straighter, faster and smoother movements when the decoder predicts the user's future intent. We also characterize how BMI performance changes as a function of delay, and explore offline how the accuracy of future prediction decoders varies at different time leads. Significance. This study is the first to characterize the effects of control delays in a BMI and to show that decoding the user's future intent can compensate for the negative effect of control delay on BMI performance.

/pdf/improving-brain-machine-interface-performance-by-decoding-138e37fkx2.pdf

Improving brain?machine interface performance by decoding intended future movements

One of the main challenges in the field of robotics is to build machines that can function intelligently in unstructured environments. Because of this, the field has witnessed a trend away from the sense-plan-act paradigm where the robot makes an attempt to model everything before planning and acting. Nevertheless, few approaches to robotic grasping and manipulation have been proposed that do not require detailed geometric models of the manipulation environment. One exception is the control-based approach where closed-loop controllers reactively generate grasping and manipulation behavior. This thesis develops and extends the control-based approach to grasping and manipulation and proposes a new framework for learning control-based skills based on generalized solutions. 
This thesis extends control-based approaches to grasping and manipulation in several ways. First, several new controllers relevant to reaching and grasping are proposed, including a grasp controller that slides contacts over the surface of an object toward good grasp configurations by using haptic feedback. The number of different grasps that can be generated using grasp controllers is expanded through the use of virtual contacts. In addition, a new approach to statically-stable dexterous manipulation is proposed whereby the robot navigates through a space of statically stable grasp configurations by executing closed-loop controllers. In a series of experiments, grasp controllers are shown to be a practical approach to synthesizing different grasps from a variety of different starting configurations. This thesis also proposes a new approach to learning control-based behaviors by applying a generalized solution in new situations. Instead of searching the entire space of all controller sequences and combinations, only variations of a generalized solution, encoded by an action schema, are considered. A new algorithm, known as schema structured learning, is proposed that learns how to apply the generalized solution in different problem contexts through a process of trial and error. This approach is applied to the grasp synthesis problem, enabling a robot to learn grasp skills with relatively little training experience. The algorithm learns to select an appropriate reach-grasp strategy based on coarse visual context. In an experiment where a dexterous humanoid robot grasps a range of grocery items it had no prior experience with, the learned grasp skills are shown to generalize well to new objects and object configurations.

Andrew H. Fagg

Papers

Null-Space Grasp Control: Theory and Experiments

Manipulation gaits: sequences of grasp control tasks

Tracing patterns and attention: humanoid robot cognition

Improving brain?machine interface performance by decoding intended future movements

Learning and generalizing control-based grasping and manipulation skills