Learning new tricks: Corticostriatal dynamics during novel skill learning

Learning novel actions and skills is a prevalent ability across multiple species and a critical feature for survival and competence in a constantly changing world. Novel actions are generated and learned through a process of trial and error, where an animal explores the environment around itself, generates multiple patterns of behavior and selects the ones that increase the likelihood of positive outcomes. Proper adaptation and execution of the selected behavior requires the coordination of several biomechanical features by the animal. Cortico-basal ganglia circuits and loops are critically involved in the acquisition, learning and consolidation of motor skills.(...)

Management consulting labs-Galp loyalty learning experience

The purpose of this project was to analyze Galp’s loyalty approach in the Portuguese fuel market given the industry context, namely the entry of hypermarket and the resulting increase in competitiveness. The team performed analyses based on analytical models, qualitative research and internal interviews in order to assess Galp’s potential in the field of loyalty and consumers’ behavior. The final recommendations were based on incremental improvements to the Galp’s existing loyalty tool and an innovative paradigm change of the approach to loyalty.

A behavioral study of alpha-1b adrenergic receptor knockout mice: increased reaction to novelty and selectively reduced learning capacities.

Knockout mice lacking the alpha-1b adrenergic receptor were tested in behavioral experiments. Reaction to novelty was first assessed in a simple test in which the time taken by the knockout mice and their littermate controls to enter a second compartment was compared. Then the mice were tested in an open field to which unknown objects were subsequently added. Special novelty was introduced by moving one of the familiar objects to another location in the open field. Spatial behavior and memory were further studied in a homing board test, and in the water maze. The alpha-1b knockout mice showed an enhanced reactivity to new situations. They were faster to enter the new environment, covered longer paths in the open field, and spent more time exploring the new objects. They reacted like controls to modification inducing spatial novelty. In the homing board test, both the knockout mice and the control mice seemed to use a combination of distant visual and proximal olfactory cues, showing place preference only if the two types of cues were redundant. In the water maze the alpha-1b knockout mice were unable to learn the task, which was confirmed in a probe trial without platform. They were perfectly able, however, to escape in a visible platform procedure. These results confirm previous findings showing that the noradrenergic pathway is important for the modulation of behaviors such as reaction to novelty and exploration, and suggest that this is mediated, at least partly, through the alpha-1b adrenergic receptors. The lack of alpha-1b adrenergic receptors in spatial orientation does not seem important in cue-rich tasks but may interfere with orientation in situations providing distant cues only.

Olfactory cues potentiate learning of distant visuospatial information

The influence of proximal olfactory cues on place learning and memory was tested in two different spatial tasks. Rats were trained to find a hole leading to their home cage or a single food source in an array of petri dishes. The two apparatuses differed both by the type of reinforcement (return to the home cage or food reward) and the local characteristics of the goal (masked holes or salient dishes). In both cases, the goal was in a fixed location relative to distant visual landmarks and could be marked by a local olfactory cue. Thus, the position of the goal was defined by two sets of redundant cues, each of which was sufficient to allow the discrimination of the goal location. These experiments were conducted with two strains of hooded rats (Long-Evans and PVG), which show different speeds of acquisition in place learning tasks. They revealed that the presence of an olfactory cue marking the goal facilitated learning of its location and that the facilitation persisted after the removal of the cue. Thus, the proximal olfactory cue appeared to potentiate learning and memory of the goal location relative to distant environmental cues. This facilitating effect was only detected when the expression of spatial memory was not already optimal, i.e., during the early phase of acquisition. It was not limited to a particular strain.

Spatial relational learning and memory abilities do not differ between men and women in a real-world, open-field environment

This study assesses gender differences in spatial and non-spatial relational learning and memory in adult humans behaving freely in a real-world, open-field environment. In Experiment 1, we tested the use of proximal landmarks as conditional cues allowing subjects to predict the location of rewards hidden in one of two sets of three distinct locations. Subjects were tested in two different conditions: (1) when local visual cues marked the potentially-rewarded locations, and (2) when no local visual cues marked the potentially-rewarded locations. We found that only 17 of 20 adults (8 males, 9 females) used the proximal landmarks to predict the locations of the rewards. Although females exhibited higher exploratory behavior at the beginning of testing, males and females discriminated the potentially-rewarded locations similarly when local visual cues were present. Interestingly, when the spatial and local information conflicted in predicting the reward locations, males considered both spatial and local information, whereas females ignored the spatial information. However, in the absence of local visual cues females discriminated the potentially-rewarded locations as well as males. In Experiment 2, subjects (9 males, 9 females) were tested with three asymmetrically-arranged rewarded locations, which were marked by local cues on alternate trials. Again, females discriminated the rewarded locations as well as males in the presence or absence of local cues. In sum, although particular aspects of task performance might differ between genders, we found no evidence that women have poorer allocentric spatial relational learning and memory abilities than men in a real-world, open-field environment.

Social sensing for psychology: Automated interpersonal behavior assessment

In this article, we show how the use of state-of-the-art methods in computer science based on machine perception and learning allows the unobtrusive capture and automated analysis of interpersonal behavior in real time (social sensing). Given the high ecological validity of the behavioral sensing, the ease of behavioral-cue extraction for large groups over long observation periods in the field, the possibility of investigating completely new research questions, and the ability to provide people with immediate feedback on behavior, social sensing will fundamentally impact psychology.

A school-based cognitive behavior modification intervention strategy for learned helpless mentally retarded students

This study assessed the usefulness of a cognitive behavior modification (CBM) intervention package with mentally retarded students in overcoming learned helplessness and improving learning strategies. It also examined the feasibility of instructing teachers in the use of such a training program for a classroom setting. A modified single subject design across individuals was employed using two groups of three subjects. Three students from each of two segregated schools for the mentally retarded were selected using a teacher questionnaire and pupil checklist of the most learned helpless students enrolled there. Three additional learned helplessness assessments were conducted on each subject before and after the intervention in order to evaluate the usefulness of the program in alleviating learned helplessness. A classroom environment was created with the three students from each school engaged in three twenty minute work sessions a week with the experimenter and a tutor experimenter (TE) as instructors. Baseline measurements were established on seven targeted behaviors for each subject: task-relevant speech, task-irrelevant speech, speech denoting a positive evaluation of performance, speech denoting a negative evaluation of performance, proportion of time on task, non-verbal positive evaluation of performance and non-verbal negative evaluation of performance. The intervention package combined a variety of CBM techniques such as Meichenbaum's (1977) Stop, Look and Listen approach, role rehearsal and feedback. During the intervention each subject met with his TE twice a week for an individual half-hour session and one joint twenty minute session with all three students, the experimentor and one TE. Five weeks after the end of this experiment one follow up probe was conducted. All baseline, post-intervention and probe sessions were videotaped. The seven targeted behaviors were coded and comparisons of baseline, post intervention, and probe testing were presented in graph form. Results showed a reduction in learned helplessness in all subjects. Improvement was noted in each of the seven targeted behaviors for each of the six subjects. This study indicated that mentally retarded children can be taught to reduce learned helplessness with the aid of a CBM intervention package. It also showed that CBM is a viable approach in helping mentally retarded students acquire more effective learning strategies. Because the TEs (Tutor experimenters) had no trouble learning and implementing this program, it was considered feasible for teachers to use similar methods in the classroom.

Towards Non Invasive Diagnosis of Scoliosis Using Semi-supervised Learning Approach

In this paper, a new methodology for the prediction of scoliosis curve types from non invasive acquisitions of the back surface of the trunk is proposed. One hundred and fifty-nine scoliosis patients had their back surface acquired in 3D using an optical digitizer. Each surface is then characterized by 45 local measurements of the back surface rotation. Using a semi-supervised algorithm, the classifier is trained with only 32 labeled and 58 unlabeled data. Tested on 69 new samples, the classifier succeeded in classifying correctly 87.0% of the data. After reducing the number of labeled training samples to 12, the behavior of the resulting classifier tends to be similar to the reference case where the classifier is trained only with the maximum number of available labeled data. Moreover, the addition of unlabeled data guided the classifier towards more generalizable boundaries between the classes. Those results provide a proof of feasibility for using a semi-supervised learning algorithm to train a classifier for the prediction of a scoliosis curve type, when only a few training data are labeled. This constitutes a promising clinical finding since it will allow the diagnosis and the follow-up of scoliotic deformities without exposing the patient to X-ray radiations.

Reinforcement Learning by Policy Search

One objective of artificial intelligence is to model the behavior of an intelligent agent interacting with its environment. The environment's transformations can be modeled as a Markov chain, whose state is partially observable to the agent and affected by its actions; such processes are known as partially observable Markov decision processes (POMDPs). While the environment's dynamics are assumed to obey certain rules, the agent does not know them and must learn. In this dissertation we focus on the agent's adaptation as captured by the reinforcement learning framework. This means learning a policy---a mapping of observations into actions---based on feedback from the environment. The learning can be viewed as browsing a set of policies while evaluating them by trial through interaction with the environment. The set of policies is constrained by the architecture of the agent's controller. POMDPs require a controller to have a memory. We investigate controllers with memory, including controllers with external memory, finite state controllers and distributed controllers for multi-agent systems. For these various controllers we work out the details of the algorithms which learn by ascending the gradient of expected cumulative reinforcement. Building on statistical learning theory and experiment design theory, a policy evaluation algorithm is developed for the case of experience re-use. We address the question of sufficient experience for uniform convergence of policy evaluation and obtain sample complexity bounds for various estimators. Finally, we demonstrate the performance of the proposed algorithms on several domains, the most complex of which is simulated adaptive packet routing in a telecommunication network.

Teaching an Old Robot New Tricks: Learning Novel Tasks via Interaction with People and Things

As AI has begun to reach out beyond its symbolic, objectivist roots into the embodied, experientialist realm, many projects are exploring different aspects of creating machines which interact with and respond to the world as humans do. Techniques for visual processing, object recognition, emotional response, gesture production and recognition, etc., are necessary components of a complete humanoid robot. However, most projects invariably concentrate on developing a few of these individual components, neglecting the issue of how all of these pieces would eventually fit together. The focus of the work in this dissertation is on creating a framework into which such specific competencies can be embedded, in a way that they can interact with each other and build layers of new functionality. To be of any practical value, such a framework must satisfy the real-world constraints of functioning in real-time with noisy sensors and actuators. The humanoid robot Cog provides an unapologetically adequate platform from which to take on such a challenge. This work makes three contributions to embodied AI. First, it offers a general-purpose architecture for developing behavior-based systems distributed over networks of PC's. Second, it provides a motor-control system that simulates several biological features which impact the development of motor behavior. Third, it develops a framework for a system which enables a robot to learn new behaviors via interacting with itself and the outside world. A few basic functional modules are built into this framework, enough to demonstrate the robot learning some very simple behaviors taught by a human trainer. A primary motivation for this project is the notion that it is practically impossible to build an "intelligent" machine unless it is designed partly to build itself. This work is a proof-of-concept of such an approach to integrating multiple perceptual and motor systems into a complete learning agent.

Learning Commonsense Categorical Knowledge in a Thread Memory System

If we are to understand how we can build machines capable of broad purpose learning and reasoning, we must first aim to build systems that can represent, acquire, and reason about the kinds of commonsense knowledge that we humans have about the world. This endeavor suggests steps such as identifying the kinds of knowledge people commonly have about the world, constructing suitable knowledge representations, and exploring the mechanisms that people use to make judgments about the everyday world. In this work, I contribute to these goals by proposing an architecture for a system that can learn commonsense knowledge about the properties and behavior of objects in the world. The architecture described here augments previous machine learning systems in four ways: (1) it relies on a seven dimensional notion of context, built from information recently given to the system, to learn and reason about objects' properties; (2) it has multiple methods that it can use to reason about objects, so that when one method fails, it can fall back on others; (3) it illustrates the usefulness of reasoning about objects by thinking about their similarity to other, better known objects, and by inferring properties of objects from the categories that they belong to; and (4) it represents an attempt to build an autonomous learner and reasoner, that sets its own goals for learning about the world and deduces new facts by reflecting on its acquired knowledge. This thesis describes this architecture, as well as a first implementation, that can learn from sentences such as ``A blue bird flew to the tree'' and ``The small bird flew to the cage'' that birds can fly. One of the main contributions of this work lies in suggesting a further set of salient ideas about how we can build broader purpose commonsense artificial learners and reasoners.

A Formulation for Active Learning with Applications to Object Detection

We discuss a formulation for active example selection for function learning problems. This formulation is obtained by adapting Fedorov's optimal experiment design to the learning problem. We specifically show how to analytically derive example selection algorithms for certain well defined function classes. We then explore the behavior and sample complexity of such active learning algorithms. Finally, we view object detection as a special case of function learning and show how our formulation reduces to a useful heuristic to choose examples to reduce the generalization error.

Interaction and Intelligent Behavior

We introduce basic behaviors as primitives for control and learning in situated, embodied agents interacting in complex domains. We propose methods for selecting, formally specifying, algorithmically implementing, empirically evaluating, and combining behaviors from a basic set. We also introduce a general methodology for automatically constructing higher--level behaviors by learning to select from this set. Based on a formulation of reinforcement learning using conditions, behaviors, and shaped reinforcement, out approach makes behavior selection learnable in noisy, uncertain environments with stochastic dynamics. All described ideas are validated with groups of up to 20 mobile robots performing safe--wandering, following, aggregation, dispersion, homing, flocking, foraging, and learning to forage.

Hybrid coordination of reinforcement learning-based behaviors for AUV control

This paper proposes a hybrid coordination method for behavior-based control architectures. The hybrid method takes advantages of the robustness and modularity in competitive approaches as well as optimized trajectories in cooperative ones. This paper shows the feasibility of applying this hybrid method with a 3D-navigation to an autonomous underwater vehicle (AUV). The behaviors are learnt online by means of reinforcement learning. A continuous Q-learning implemented with a feed-forward neural network is employed. Realistic simulations were carried out. The results obtained show the good performance of the hybrid method on behavior coordination as well as the convergence of the behaviors

Efficient learning of reactive robot behaviors with a Neural-Q_learning approach

The purpose of this paper is to propose a Neural-Q_learning approach designed for online learning of simple and reactive robot behaviors. In this approach, the Q_function is generalized by a multi-layer neural network allowing the use of continuous states and actions. The algorithm uses a database of the most recent learning samples to accelerate and guarantee the convergence. Each Neural-Q_learning function represents an independent, reactive and adaptive behavior which maps sensorial states to robot control actions. A group of these behaviors constitutes a reactive control scheme designed to fulfill simple missions. The paper centers on the description of the Neural-Q_learning based behaviors showing their performance with an underwater robot in a target following task. Real experiments demonstrate the convergence and stability of the learning system, pointing out its suitability for online robot learning. Advantages and limitations are discussed