Structure learning of graphical models for task-oriented robot grasping

In the collective imaginaries a robot is a human like machine as any androids in science fiction. However the type of robots that you will encounter most frequently are machinery that do work that is too dangerous, boring or onerous. Most of the robots in the world are of this type. They can be found in auto, medical, manufacturing and space industries. Therefore a robot is a system that contains sensors, control systems, manipulators, power supplies and software all working together to perform a task. The development and use of such a system is an active area of research and one of the main problems is the development of interaction skills with the surrounding environment, which include the ability to grasp objects. To perform this task the robot needs to sense the environment and acquire the object informations, physical attributes that may influence a grasp. Humans can solve this grasping problem easily due to their past experiences, that is why many researchers are approaching it from a machine learning perspective finding grasp of an object using information of already known objects. But humans can select the best grasp amongst a vast repertoire not only considering the physical attributes of the object to grasp but even to obtain a certain effect. This is why in our case the study in the area of robot manipulation is focused on grasping and integrating symbolic tasks with data gained through sensors. The learning model is based on Bayesian Network to encode the statistical dependencies between the data collected by the sensors and the symbolic task. This data representation has several advantages. It allows to take into account the uncertainty of the real world, allowing to deal with sensor noise, encodes notion of causality and provides an unified network for learning. Since the network is actually implemented and based on the human expert knowledge, it is very interesting to implement an automated method to learn the structure as in the future more tasks and object features can be introduced and a complex network design based only on human expert knowledge can become unreliable. Since structure learning algorithms presents some weaknesses, the goal of this thesis is to analyze real data used in the network modeled by the human expert, implement a feasible structure learning approach and compare the results with the network designed by the expert in order to possibly enhance it.

How Porous are the Walls that Separate Us?: Transformative Service-Learning, Women’s Incarceration, and the Unsettled Self

In this article, we refine a politics of thinking from the margins by exploring a pedagogical model that advances transformative notions of service learning as social justice teaching. Drawing on a recent course we taught involving both incarcerated women and traditional college students, we contend that when communication among differentiated and stratified parties occurs, one possible result is not just a view of the other but also a transformation of the self and other. More specifically, we suggest that an engaged feminist praxis of teaching incarcerated women together with college students helps illuminate the porous nature of fixed markers that purport to reveal our identities (e.g., race and gender), to emplace our bodies (e.g., within institutions, prison gates, and walls), and to specify our locations (e.g., cultural, geographic, socialeconomic). One crucial theoretical insight our work makes clear is that the model of social justice teaching to which we aspired necessitates re-conceptualizing ourselves as students and professors whose subjectivities are necessarily relational and emergent.

Learned irrelevance and associative learning is attenuated in individuals at risk for psychosis but not in asymptomatic first-degree relatives of schizophrenia patients: translational state markers of psychosis?

Learned irrelevance (LIrr) refers to a form of selective learning that develops as a result of prior noncorrelated exposures of the predicted and predictor stimuli. In learning situations that depend on the associative link between the predicted and predictor stimuli, LIrr is expressed as a retardation of learning. It represents a form of modulation of learning by selective attention. Given the relevance of selective attention impairment to both positive and cognitive schizophrenia symptoms, the question remains whether LIrr impairment represents a state (relating to symptom manifestation) or trait (relating to schizophrenia endophenotypes) marker of human psychosis. We examined this by evaluating the expression of LIrr in an associative learning paradigm in (1) asymptomatic first-degree relatives of schizophrenia patients (SZ-relatives) and in (2) individuals exhibiting prodromal signs of psychosis ("ultrahigh risk" [UHR] patients) in each case relative to demographically matched healthy control subjects. There was no evidence for aberrant LIrr in SZ-relatives, but LIrr as well as associative learning were attenuated in UHR patients. It is concluded that LIrr deficiency in conjunction with a learning impairment might be a useful state marker predictive of psychotic state but a relatively weak link to a potential schizophrenia endophenotype.

Dopamine-related deficit in reward learning after catecholamine depletion in unmedicated, remitted subjects with bulimia nervosa

Disturbances in reward processing have been implicated in bulimia nervosa (BN). Abnormalities in processing reward-related stimuli might be linked to dysfunctions of the catecholaminergic neurotransmitter system, but findings have been inconclusive. A powerful way to investigate the relationship between catecholaminergic function and behavior is to examine behavioral changes in response to experimental catecholamine depletion (CD). The purpose of this study was to uncover putative catecholaminergic dysfunction in remitted subjects with BN who performed a reinforcement-learning task after CD. CD was achieved by oral alpha-methyl-para-tyrosine (AMPT) in 19 unmedicated female subjects with remitted BN (rBN) and 28 demographically matched healthy female controls (HC). Sham depletion administered identical capsules containing diphenhydramine. The study design consisted of a randomized, double-blind, placebo-controlled crossover, single-site experimental trial. The main outcome measures were reward learning in a probabilistic reward task analyzed using signal-detection theory. Secondary outcome measures included self-report assessments, including the Eating Disorder Examination-Questionnaire. Relative to healthy controls, rBN subjects were characterized by blunted reward learning in the AMPT-but not in placebo-condition. Highlighting the specificity of these findings, groups did not differ in their ability to perceptually distinguish between stimuli. Increased CD-induced anhedonic (but not eating disorder) symptoms were associated with a reduced response bias toward a more frequently rewarded stimulus. In conclusion, under CD, rBN subjects showed reduced reward learning compared with healthy control subjects. These deficits uncover disturbance of the central reward processing systems in rBN related to altered brain catecholamine levels, which might reflect a trait-like deficit increasing vulnerability to BN.

Energy-Economical Heuristically Based Control of Compass Gait Walking on Stochastically Varying Terrain

Investigation uses simulation to explore the inherent tradeoffs ofcontrolling high-speed and highly robust walking robots while minimizing energy consumption. Using a novel controller which optimizes robustness, energy economy, and speed of a simulated robot on rough terrain, the user can adjust their priorities between these three outcome measures and systematically generate a performance curveassessing the tradeoffs associated with these metrics.

Maternal and Postweaning High-Fat Diets Disturb Hippocampal Gene Expression, Learning, and Memory Function

We tested the hypothesis that excess saturated fat consumption during pregnancy, lactation, and/or postweaning alters the expression of genes mediating hippocampal synaptic efficacy and impairs spatial learning and memory in adulthood. Dams were fed control chow or a diet high in saturated fat before mating, during pregnancy, and into lactation. Offspring were weaned to either standard chow or a diet high in saturated fat. The Morris Water Maze was used to evaluate spatial learning and memory. Open field testing was used to evaluate motor activity. Hippocampal gene expression in adult males was measured using RT-PCR and ELISA. Offspring from high fat-fed dams took longer, swam farther, and faster to try and find the hidden platform during the 5-day learning period. Control offspring consuming standard chow spent the most time in memory quadrant during the probe test. Offspring from high fat-fed dams consuming excess saturated fat spent the least. The levels of mRNA and protein for brain-derived neurotrophic factor and activity-regulated cytoskeletal-associated protein were significantly decreased by maternal diet effects. Nerve growth factor mRNA and protein levels were significantly reduced in response to both maternal and postweaning high-fat diets. Expression levels for the N-methyl-D-aspartate receptor (NMDA) receptor subunit NR2B as well as synaptophysin were significantly decreased in response to both maternal and postweaning diets. Synaptotagmin was significantly increased in offspring from high fat-fed dams. These data support the hypothesis that exposure to excess saturated fat during hippocampal development is associated with complex patterns of gene expression and deficits in learning and memory.

First clinical trial of tomographic neurofeedback in attention-deficit/hyperactivity disorder: evaluation of voluntary cortical control

Tomographic neurofeedback (tNF) training was evaluated as a treatment for attention-deficit/hyperactivity disorder (ADHD). To investigate the specificity of the treatment, outcomes were related to learning during tNF.

Functional neuroimaging of emotional learning and autonomic reactions

This article provides a selective overview of the functional neuroimaging literature with an emphasis on emotional activation processes. Emotions are fast and flexible response systems that provide basic tendencies for adaptive action. From the range of involved component functions, we first discuss selected automatic mechanisms that control basic adaptational changes. Second, we illustrate how neuroimaging work has contributed to the mapping of the network components associated with basic emotion families (fear, anger, disgust, happiness), and secondary dimensional concepts that organise the meaning space for subjective experience and verbal labels (emotional valence, activity/intensity, approach/withdrawal, etc.). Third, results and methodological difficulties are discussed in view of own neuroimaging experiments that investigated the component functions involved in emotional learning. The amygdala, prefrontal cortex, and striatum form a network of reciprocal connections that show topographically distinct patterns of activity as a correlate of up and down regulation processes during an emotional episode. Emotional modulations of other brain systems have attracted recent research interests. Emotional neuroimaging calls for more representative designs that highlight the modulatory influences of regulation strategies and socio-cultural factors responsible for inhibitory control and extinction. We conclude by emphasising the relevance of the temporal process dynamics of emotional activations that may provide improved prediction of individual differences in emotionality.

When conditioned responses "fire back": bidirectional cross-activation creates learning opportunities in synesthesia

In grapheme-color synesthesia, the letter "c" printed in black may be experienced as red, but typically the color red does not trigger the experience of the letter "c." Therefore, at the level of subjective experience, cross-activation is usually unidirectional. However, recent evidence from digit-color synesthesia suggests that at an implicit level bidirectional cross-activation can occur. Here we demonstrate that this finding is not restricted to this specific type of synesthesia. We introduce a new method that enables the investigation of bidirectionality in other types of synesthesia. We found that a group of grapheme-color synesthetes, but not a control group, showed a startle in response to a color-inducing grapheme after a startle response was conditioned to the specific corresponding color. These results implicate that when the startle response was associated with the real color an association between shock and the grapheme was also established. By this mechanism (i.e. implicit cross-activation) the conditioned response to the real color generalized to the synesthetic color. We suggest that parietal brain areas are responsible for this neural backfiring.

The free radical scavenger alpha-phenyl-tert-butyl nitrone aggravates hippocampal apoptosis and learning deficits in experimental pneumococcal meningitis

The effect of adjuvant therapy with the radical scavenger alpha-phenyl-tert-butyl nitrone (PBN; 100 mg/kg given intraperitoneally every 8 h for 5 days) on brain injury and learning function was evaluated in an infant rat model of pneumococcal meningitis. Meningitis led to cortical necrotic injury (median, 3.97% [range, 0%-38.9%] of the cortex), which was reduced to a median of 0% (range, 0%-30.9%) of the cortex (P<.001) by PBN. However, neuronal apoptosis in the hippocampal dentate gyrus was increased by PBN, compared with that by saline (median score, 1.15 [range, 0.04-1.73] vs. 0.31 [range, 0-0.92]; P<.001). Learning function 3 weeks after cured infection, as assessed by the Morris water maze, was decreased, compared with that in uninfected control animals (P<.001). Parallel to the increase in hippocampal apoptosis, PBN further impaired learning in infected animals, compared with that in saline-treated animals (P<.02). These results contrast with those of an earlier study, in which PBN reduced cortical and hippocampal neuronal injury in group B streptococcal meningitis. Thus, in pneumococcal meningitis, antioxidant therapy with PBN aggravates hippocampal injury and learning deficits.

Improving learning by connecting chemistry curriculum to students' experiences

The purpose of the study was to design, implement, and assess the effects of a teaching unit about fuel sources and chemical energy on students’ learning. The unit was designed to incorporate students’ experiences in a way that was aligned with the Michigan High School Content Expectations. The study was completed with all of the students taking General Chemistry in a rural Michigan high school in the 2010-11 school year. There were 138 participants total. The participants were mostly Caucasian and the majority were in the 11th grade. Of these, 77 constituted the experimental group and were taught the unit. The additional 61 participants in the control group were given the posttest only. Data was derived from the results of pre/post tests, final assessment projects, and the researcher’s observations. A pretest that contained questions about the fuel sources was administered at the beginning of the unit. An identical posttest was administered at the completion of the unit. A final assessment project required students to choose the best fuel source for the area, and support their opinion with facts and data from their research or the learning activities and labs performed in class. The results of the study revealed that the teaching unit did produce significant learning gains in the experimental group. The results also indicated that the teaching unit added value to the current General Chemistry curriculum by expanding what students were learning. The instructional goals of the unit were aligned with the Michigan High School Content Expectations. The results also revealed that the students were able to learn to support their thinking and decisions with explanations based on the data and labs. These are essential science literacy skills. The study supported the view that connecting the required curriculum with students’ experiences and interests was effective, and that students can learn important science literacy skills, such as providing support for arguments and communicating scientific explanations, when given adequate teacher support.

Interference during the implicit learning of two different motor sequences

It has been demonstrated that learning a second motor task after having learned a first task may interfere with the long-term consolidation of the first task. However, little is known about immediate changes in the representation of the motor memory in the early acquisition phase within the first minutes of the learning process. Therefore, we investigated such early interference effects with an implicit serial reaction time task in 55 healthy subjects. Each subject performed either a sequence learning task involving two different sequences, or a random control task. The results showed that learning the first sequence led to only a slight, short-lived interference effect in the early acquisition phase of the second sequence. Overall, learning of neither sequence was impaired. Furthermore, the two processes, sequence-unrelated task learning (i.e. general motor training) and the sequence learning itself did not appear to interfere with each other. In conclusion, although the long-term consolidation of a motor memory has been shown to be sensitive to other interfering memories, the present study suggests that the brain is initially able to acquire more than one new motor sequence within a short space of time without significant interference.

An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.