As evident from deep learning, very large models bring improvements in training dynamics and representation power. Yet, smaller models have benefits of energy efficiency and interpretability. To get the benefits from both ends of the spectrum we often encourage sparsity in the model. Unfortunately, most existing approaches do not have a controllable way to request a desired value of sparsity in an interpretable parameter. In this paper, we design a new sparse projection method for a set of vectors in order to achieve a desired average level of sparsity which is measured using the ratio of the $\ell_1$ and $\ell_2$ norms. Most existing methods project each vector individuality trying to achieve a target sparsity, hence the user has to choose a sparsity level for each vector (e.g., impose that all vectors have the same sparsity). Instead, we project all vectors together to achieve an average target sparsity, where the sparsity levels of the vectors is automatically tuned. We also propose a generalization of this projection using a new notion of weighted sparsity measured using the ratio of a weighted $\ell_1$ and the $\ell_2$ norms. These projections can be used in particular to sparsify the columns of a matrix, which we use to compute sparse nonnegative matrix factorization and to learn sparse deep networks.

We propose to reinterpret a standard discriminative classifier of p(y|x) as an energy based model for the joint distribution p(x,y). In this setting, the standard class probabilities can be easily computed as well as unnormalized values of p(x) and p(x|y). Within this framework, standard discriminative architectures may beused and the model can also be trained on unlabeled data. We demonstrate that energy based training of the joint distribution improves calibration, robustness, andout-of-distribution detection while also enabling our models to generate samplesrivaling the quality of recent GAN approaches. We improve upon recently proposed techniques for scaling up the training of energy based models and presentan approach which adds little overhead compared to standard classification training. Our approach is the first to achieve performance rivaling the state-of-the-artin both generative and discriminative learning within one hybrid model.

A BSTRACT Understanding and learning the characteristics of network paths has been of particular interest for decades and has led to several successful applications. Such analysis becomes challenging for urban networks as their size and complexity are significantly higher compared to other networks. The state-of-the-art machine learning (ML) techniques allow us to detect hidden patterns and, thus, infer the features associated with them. However, very little is known about the impact on the performance of such predictive models by the use of different input representations. In this paper, we design and evaluate six different graph input representations (i.e., representations of the network paths), by considering the network's topological and temporal characteristics, for being used as inputs for machine learning models to learn the behavior of urban networks paths. The representations are validated and then tested with a real-world taxi journeys dataset predicting the tips using a road network of New Y ork. Our results demonstrate that the input representations that use temporal information help the model to achieve the highest accuracy (RMSE of 1.42$). K eywords Urban Networks, Graph Learning, Path Representation 1 Introduction Numerous important problems can be studied using the conceptual and theoretical framework of network science. Several structure and topological properties of networks have been widely studied in the recent years ([12, 14, 5, 9]). One of the most basic concepts in network science is the definition of network path ([3, 2]), i.e., a sequence of edges that joins a sequence of edges.

With the widespread integration of AI in everyday and critical technologies, it seems inevitable to witness increasing instances of failure in AI systems. In such cases, there arises a need for technical investigations that produce legally acceptable and scientifically indisputable findings and conclusions on the causes of such failures. Inspired by the domain of cy-ber forensics, this paper introduces the need for the establishment of AI F orensics as a new discipline under AI safety. Furthermore, we propose a taxonomy of the subfields under this discipline, and present a discussion on the foundational challenges that lay ahead of this new research area. Introduction Recent advances in Artificial Intelligence (AI) have given rise to the rapidly growing adoption of such techniques by a vast array of industries and technologies.

In this paper we propose a deep learning method for performing attributed-based music-to-image translation. The proposed method is applied for synthesizing visual stories according to the sentiment expressed by songs. The generated images aim to induce the same feelings to the viewers, as the original song does, reinforcing the primary aim of music, i.e., communicating feelings. The process of music-to-image translation poses unique challenges, mainly due to the unstable mapping between the different modalities involved in this process. In this paper, we employ a trainable cross-modal translation method to overcome this limitation, leading to the first, to the best of our knowledge, deep learning method for generating sentiment-aware visual stories. Various aspects of the proposed method are extensively evaluated and discussed using different songs.

Reinforcement learning agents can include different components, such as policies, value functions, state representations, and environment models. Any or all of these can be the loci of knowledge, i.e., structures where knowledge, whether given or learned, can be deposited and reused. The objective of an agent is to behave so as to maximise the sum of a suitable scalar function of state: the reward. As far as the learning algorithm is concerned, these rewards are typically given and immutable. In this paper we instead consider the proposition that the reward function itself may be a good locus of knowledge. This is consistent with a common use, in the literature, of hand-designed intrinsic rewards to improve the learning dynamics of an agent. We adopt the multi-lifetime setting of the Optimal Rewards Framework, and propose to meta-learn an intrinsic reward function from experience that allows agents to maximise their extrinsic rewards accumulated until the end of their lifetimes. Rewards as a locus of knowledge provide guidance on "what" the agent should strive to do rather than "how" the agent should behave; the latter is more directly captured in policies or value functions for example. Thus, our focus here is on demonstrating the following: (1) that it is feasible to meta-learn good reward functions, (2) that the learned reward functions can capture interesting kinds of "what" knowledge, and (3) that because of the indirectness of this form of knowledge the learned reward functions can generalise to other kinds of agents and to changes in the dynamics of the environment. Reinforcement learning agents can store knowledge in their policies, value functions, state representations, and models of the environment dynamics. These components can be the loci of knowledge in the sense that they are structures in which knowledge, either learned from experience by the agent's algorithm or given by the agent-designer, can be deposited and reused.

Not submitted to the 30th International Conference on Automated Planning and SchedulingNeural-Symbolic Descriptive Action Model from Images: The Search for STRIPS Masataro Asai MIT -IBM Watson AI Lab, Cambridge USA IBM Research Abstract Recent work on Neural-Symbolic systems that learn the discrete planning model from images has opened a promising direction for expanding the scope of Automated Planning and Scheduling to the raw, noisy data. However, previous work only partially addressed this problem, utilizing the black-box neural model as the successor generator. In this work, we propose Double-Stage Action Model Acquisition (DSAMA), a system that obtains a descriptive PDDL action model with explicit preconditions and effects over the propositional variables unsupervised-learned from images. DSAMA trains a set of Random Forest rule-based classifiers and compiles them into logical formulae in PDDL. While we obtained a competitively accurate PDDL model compared to a black-box model, we observed that the resulting PDDL is too large and complex for the state-of-the-art standard planners such as Fast Downward primarily due to the PDDL-SAS translator bottleneck. From this negative result, we show that this translator bottleneck cannot be addressed just by using a different, existing rule-based learning method, and we point to the potential future directions. 1 Introduction Recently, Latplan system (Asai and Fukunaga 2018) successfully connected a subsymbolic neural network (NN) system and a symbolic Classical Planning system to solve various visually presented puzzle domains. The system consists of four parts: 1) The State AutoEncoder (SAE) neural network learns a bidirectional mapping between images and propositional states with unsupervised training. The proposed framework opened a promising direction for applying a variety of symbolic methods to the real world -- For example, the search space generated by Latplan was shown to be compatible with a symbolic Goal Recognition system (Amado et al. 2018a; 2018b). Several variations replacing the state encoding modules have also been proposed: Causal InfoGAN (Kurutach et al. 2018) uses a GAN-based framework, First-Order SAE (Asai 2019) learns the First Order Logic symbols (instead of the propositional ones), and Zero-Suppressed SAE (Asai (:action a0:parameters ():precondition [D0]:effect (and (when [E00] (z0)) (when (not [E00]) (not (z0))) (when [E01] (z1)) (when (not [E01]) (not (z1))) ...)) Figure 1: An example DSAMA compilation result for the first action (i.e. Despite these efforts, Latplan is missing a critical feature of the traditional Classical Planning systems: The use of State-of-the-Art heuristic functions.

This year, we have seen all the hype around AI Deep Learning. With recent innovations, deep learning demonstrated its usefulness in performing tasks such as image recognition, voice recognition, price forecasting, across many industries. It's easy to overestimate deep learning's capabilities and pretend it's the magic bullet that will allow AI to obtain General Intelligence. In truth, we are still far away from that. However, deep learning has a relatively unknown partner: Reinforcement Learning.

Online Artificial Intelligence Course with Python and TensorFlow 2.0 library launched today. Students looking for fast track course can join this training course. The Artificial Intelligence Course with TensorFlow 2.0 includes latest API released in the TensorFlow. According to the firm this course is online instructor laid training where students will be able to learn from highly qualified instructors. The demand of highly skilled data scientist is expected to grow in 2020 and beyond.

When it comes to AI (Artificial Intelligence), VCs (venture capitalists) continue to be aggressive with their fundings. During the third quarter, 965 AI-related companies in the US raised a total of $13.5 billion. In fact, this year should see a record in total fundings (last year's total came to $16.8 billion). Some of the deals have been, well, staggering. Just look at the $1 billion that Microsoft shelled out for an equity stake in OpenAI (the company is one of the few that is pursuing Strong AI). So what has been the result of all this activity? What have been the breakthroughs for AI this year?