Robots should understand both semantics and physics to be functional in the real world. While robot platforms provide means for interacting with the physical world they cannot autonomously acquire object-level semantics without needing human. In this paper, we investigate how to minimize human effort and intervention to teach robots perform real world tasks that incorporate semantics. We study this question in the context of visual servoing of mobile robots and propose DIViS, a Domain Invariant policy learning approach for collision free Visual Servoing. DIViS incorporates high level semantics from previously collected static human-labeled datasets and learns collision free servoing entirely in simulation and without any real robot data. However, DIViS can directly be deployed on a real robot and is capable of servoing to the user-specified object categories while avoiding collisions in the real world. DIViS is not constrained to be queried by the final view of goal but rather is robust to servo to image goals taken from initial robot view with high occlusions without this impairing its ability to maintain a collision free path. We show the generalization capability of DIViS on real mobile robots in more than 90 real world test scenarios with various unseen object goals in unstructured environments. DIViS is compared to prior approaches via real world experiments and rigorous tests in simulation. For supplementary videos, see: \href{https://fsadeghi.github.io/DIViS}{https://fsadeghi.github.io/DIViS}

Complex environments provide structured yet variable sensory inputs. To best exploit information from these environments, organisms must evolve the ability to correctly anticipate consequences of unknown stimuli, and act on these predictions. We propose an evolutionary path for neural networks, leading an organism from reactive behavior to simple proactive behavior and from simple proactive behavior to induction-based behavior. Through in-vitro and in-silico experiments, we define the minimal conditions necessary in a network with spike-timing dependent plasticity for the organism to go from reactive to proactive behavior. Our results support the existence of small evolutionary steps and four necessary conditions allowing embodied neural networks to evolve predictive and inductive abilities from an initial reactive strategy. We extend these conditions to more general structures.

Although deep convolutional networks have achieved improved performance in many natural language tasks, they have been treated as black boxes because they are difficult to interpret. Especially, little is known about how they represent language in their intermediate layers. In an attempt to understand the representations of deep convolutional networks trained on language tasks, we show that individual units are selectively responsive to specific morphemes, words, and phrases, rather than responding to arbitrary and uninterpretable patterns. In order to quantitatively analyze such an intriguing phenomenon, we propose a concept alignment method based on how units respond to the replicated text. We conduct analyses with different architectures on multiple datasets for classification and translation tasks and provide new insights into how deep models understand natural language.

For complex real-world systems, designing controllers are a difficult task. With the advent of neural networks as a proxy for complex function approximators, it has become popular to learn the controller directly. However, these controllers are specific to a given task and need to be relearned for a new task. Alternatively, one can learn just the model of the dynamical system and compose it with external controllers. Such a model is task (and controller) agnostic and must generalize well across the state space. This paper proposes learning a "sufficiently accurate" model of the dynamics that explicitly enforces small residual error on pre-defined parts of the state-space. We formulate task agnostic controller design for this learned model as an optimization problem with state and control constraints that is solved in an online fashion. We validate this approach in simulation using a challenging contact-based Ball-Paddle system.

Deep learning (DL) based autoencoder has shown great potential to significantly enhance the physical layer performance. In this paper, we present a DL based autoencoder for interference channel. Based on a characterization of a k-user Gaussian interference channel, where the interferences are classified as different levels from weak to very strong interferences based on a coupling parameter {\alpha}, a DL neural network (NN) based autoencoder is designed to train the data set and decode the received signals. The performance such a DL autoencoder for different interference scenarios are studied, with {\alpha} known or partially known, where we assume that {\alpha} is predictable but with a varying up to 10\% at the training stage. The results demonstrate that DL based approach has a significant capability to mitigate the effect induced by a poor signal-to-noise ratio (SNR) and a high interference-to-noise ratio (INR). However, the enhancement depends on the knowledge of {\alpha} as well as the interference levels. The proposed DL approach performs well with {\alpha} up to 10\% offset for weak interference level. For strong and very strong interference channel, the offset of {\alpha} needs to be constrained to less than 5\% and 2\%, respectively, to maintain similar performance as {\alpha} is known.

V alidation is one of the most important aspects of clustering, but most approaches have been batch methods. Recently, interest has grown in providing incremental alternatives. This paper extends the incremental cluster validity index (iCVI) family to include incremental versions of Calinski-Harabasz (iCH), I index and Pakhira-Bandyopadhyay-Maulik (iI and iPBM), Silhouette (iSIL), Negentropy Increment (iNI), Representative Cross Information Potential (irCIP) and Representative Cross Entropy (irH), and Conn Index (iConn Index). Additionally, the effect of under-and over-partitioning on the behavior of these six iCVIs, the Partition Separation (PS) index, as well as two other recently developed iCVIs (incremental Xie-Beni (iXB) and incremental Davies-Bouldin (iDB)) was examined through a comparative study. Experimental results using fuzzy adaptive resonance theory (ART)-based clustering methods showed that while evidence of most under-partitioning cases could be inferred from the behaviors of all these iCVIs, over-partitioning was found to be a more challenging scenario indicated only by the iConn Index. The expansion of incremental validity indices provides significant novel opportunities for assessing and interpreting the results of unsupervised learning. L. E. Brito da Silva is with the Applied Computational Intelligence Laboratory, Department of Electrical and Computer Engineering, Missouri University of Science and Technology, Rolla, MO 65409 USA, and also with the CAPES Foundation, Ministry of Education of Brazil, Bras ılia, DF 70040-020, Brazil (email: leonardoenzo@ieee.org). N. M. Melton is with the Applied Computational Intelligence Laboratory, Department of Electrical and Computer Engineering, Missouri University of Science and Technology, Rolla, MO 65409 USA (email: niklasmelton@ieee.org). D. C. Wunsch II is with the Applied Computational Intelligence Laboratory, Department of Electrical and Computer Engineering, Missouri University of Science and Technology, Rolla, MO 65409 USA (email: wunsch@ieee.org). I NTRODUCTION Cluster validation [1] is a critical topic in cluster analysis.

We demonstrate the first possibility of a sub-linear memory sketch for solving the approximate near-neighbor search problem. In particular, we develop an online sketching algorithm that can compress $N$ vectors into a tiny sketch consisting of small arrays of counters whose size scales as $O(N^{b}\log^2{N})$, where $b < 1$ depending on the stability of the near-neighbor search. This sketch is sufficient to identify the top-$v$ near-neighbors with high probability. To the best of our knowledge, this is the first near-neighbor search algorithm that breaks the linear memory ($O(N)$) barrier. We achieve sub-linear memory by combining advances in locality sensitive hashing (LSH) based estimation, especially the recently-published ACE algorithm, with compressed sensing and heavy hitter techniques. We provide strong theoretical guarantees; in particular, our analysis sheds new light on the memory-accuracy tradeoff in the near-neighbor search setting and the role of sparsity in compressed sensing, which could be of independent interest. We rigorously evaluate our framework, which we call RACE (Repeated ACE) data structures on a friend recommendation task on the Google plus graph with more than 100,000 high-dimensional vectors. RACE provides compression that is orders of magnitude better than the random projection based alternative, which is unsurprising given the theoretical advantage. We anticipate that RACE will enable both new theoretical perspectives on near-neighbor search and new methodologies for applications like high-speed data mining, internet-of-things (IoT), and beyond.

They can be basically defined as a multi-layer stack of algorithms or modules able to gradually learn a huge number of parameters in an architecture composed by multiple nonlinear transformations (LeCun et al., 2015). Typically, and for historical reasons, a structure for deep learning is identified with advanced neural networks: deep Feed Forward, Recurrent, Auto-encoder, Convolution neural networks are very effective and used algorithms of deep learning (Schmidhuber, 2015). They demonstrated to be particularly successful in supervised classification problems arising in several fields such as image and speech recognition, gene expression data, topic classification. When the aim is uncovering unknown classes in a unsupervised classification perspective, important methods of deep learning have been developed along the lines of mixture modeling, because of their ability to decompose a heterogeneous collection of units into a finite number of subgroups with homogeneous structures (Fraley and Raftery, 2002; McLachlan and Peel, 2000). In this direction, van den Oord and Schrauwen (2014) proposed Multilayer Gaussian Mixture Models for modeling natural images; Tang et al. (2012) defined deep mixture of factor analyzers with a greedy layer-wise learning algorithm able to learn each layer at a time. Viroli and McLachlan (2019) developed a general framework for Deep Gaussian mixture models that generalizes and encompasses the previous strategies and several flexible model-based clustering methods such as mixtures of mixture models (Li, 2005), mixtures of Factor Analyzers (McLachlan et al., 2003), mixtures of factor analyzers with common factor loadings (Baek et al., 2010), heteroscedastic factor mixture analysis (Montanari and Viroli, 2010) and mixtures of factor mixture analyzers introduced by Viroli (2010). A general'take-home-message' coming from the existing deep clustering strategies is that deep methods vs shallow ones appear to be very efficient and powerful tools especially for complex high-dimensional data; on the contrary, for simple and small data structures, a deep learning strategy cannot improve performance of simpler and conventional methods or, to better say, it is like to use a'sledgehammer to crack a nut'. The motivating problem behind this work derives from ticket data (i.e.

The performance of policy gradient methods is sensitive to hyperparameter settings that must be tuned for any new application. Widely used grid search methods for tuning hyperparameters are sample inefficient and computationally expensive. More advanced methods like Population Based Training (Jaderberg et al., 2017) that learn optimal schedules for hyperparameters instead of fixed settings canyield better results, but are also sample inefficient and computationally expensive. In this paper, we propose Hyperparameter Optimisation on the Fly (HOOF), a gradient-free meta-learning algorithm that can automatically learn an optimal schedule for hyperparameters that affect the policy updatedirectly through the gradient. The main idea is to use existing trajectories sampled by the policy gradient method to optimise a one-step improvement objective,yielding a sample and computationally efficientalgorithm that is easy to implement. Our experimental results across multiple domains and algorithms show that using HOOF to learn these hyperparameter schedules leads to faster learning with improved performance.

The objective of this research is to enhance performance of Stochastic Gradient Descent (SGD) algorithm in text classification. In our research, we proposed using SGD learning with Grid-Search approach to fine-tuning hyper-parameters in order to enhance the performance of SGD classification. We explored different settings for representation, transformation and weighting features from the summary description of terrorist attacks incidents obtained from the Global Terrorism Database as a pre-classification step, and validated SGD learning on Support Vector Machine (SVM), Logistic Regression and Perceptron classifiers by stratified 10-K-fold cross-validation to compare the performance of different classifiers embedded in SGD algorithm. The research concludes that using a grid-search to find the hyper-parameters optimize SGD classification, not in the pre-classification settings only, but also in the performance of the classifiers in terms of accuracy and execution time.