Joint extraction of entities and relations has received significant attention due to its potential of providing higher performance for both tasks. Among existing methods, CopyRE is effective and novel, which uses a sequence-to-sequence framework and copy mechanism to directly generate the relation triplets. However, it suffers from two fatal problems. The model is extremely weak at differing the head and tail entity, resulting in inaccurate entity extraction. It also cannot predict multi-token entities (e.g. \textit{Steven Jobs}). To address these problems, we give a detailed analysis of the reasons behind the inaccurate entity extraction problem, and then propose a simple but extremely effective model structure to solve this problem. In addition, we propose a multi-task learning framework equipped with copy mechanism, called CopyMTL, to allow the model to predict multi-token entities. Experiments reveal the problems of CopyRE and show that our model achieves significant improvement over the current state-of-the-art method by 9% in NYT and 16% in WebNLG (F1 score). Our code is available at https://github.com/WindChimeRan/CopyMTL

In response to the continuing research interest in computational semantic analysis, we have proposed a new task for SemEval-2010: multi-way classification of mutually exclusive semantic relations between pairs of nominals. The task is designed to compare different approaches to the problem and to provide a standard testbed for future research. In this paper, we define the task, describe the creation of the datasets, and discuss the results of the participating 28 systems submitted by 10 teams.

In this paper, we describe SemEval-2013 Task 4: the definition, the data, the evaluation and the results. The task is to capture some of the meaning of English noun compounds via paraphrasing. Given a two-word noun compound, the participating system is asked to produce an explicitly ranked list of its free-form paraphrases. The list is automatically compared and evaluated against a similarly ranked list of paraphrases proposed by human annotators, recruited and managed through Amazon's Mechanical Turk. The comparison of raw paraphrases is sensitive to syntactic and morphological variation. The "gold" ranking is based on the relative popularity of paraphrases among annotators. To make the ranking more reliable, highly similar paraphrases are grouped, so as to downplay superficial differences in syntax and morphology. Three systems participated in the task. They all beat a simple baseline on one of the two evaluation measures, but not on both measures. This shows that the task is difficult.

Tel.: 79114005773 Received: date; Accepted: date; Published: date Abstract: The study presents an oscillator circuit for a spike neural network with the possibility of firing rate coding a nd sigmoid - like activation function . The circuit contains a switching element with an S - shaped current - voltage characteristic and two capacitors; one of the capacitors is shunted by a control resistor. The circuit is characterised by a strong dependence of t he frequency of relaxation oscillations on the magnitude of the control resistor. The dependence has a sigmoid - like form and we present an analytical method for dependence calculation. Finally, we describe t he concept of the spike neural network architectu re with firing rate coding based on the presented circuit for creating neuromorphic devices and artificial intelligence.

Imitation learning has gained immense popularity because of its high sample-efficiency. However, in real-world scenarios, where the trajectory distribution of most of the tasks dynamically shifts, model fitting on continuously aggregated data alone would be futile. In some cases, the distribution shifts, so much, that it is difficult for an agent to infer the new task. We propose a novel algorithm to generalize on any related task by leveraging prior knowledge on a set of specific tasks, which involves assigning importance weights to each past demonstration. We show experiments where the robot is trained from a diversity of environmental tasks and is also able to adapt to an unseen environment, using few-shot learning. We also developed a prototype robot system to test our approach on the task of visual navigation, and experimental results obtained were able to confirm these suppositions.

We present $\Gamma$-nets, a method for generalizing value function estimation over timescale. By using the timescale as one of the estimator's inputs we can estimate value for arbitrary timescales. As a result, the prediction target for any timescale is available and we are free to train on multiple timescales at each timestep. Here we empirically evaluate $\Gamma$-nets in the policy evaluation setting. We first demonstrate the approach on a square wave and then on a robot arm using linear function approximation. Next, we consider the deep reinforcement learning setting using several Atari video games. Our results show that $\Gamma$-nets can be effective for predicting arbitrary timescales, with only a small cost in accuracy as compared to learning estimators for fixed timescales. $\Gamma$-nets provide a method for compactly making predictions at many timescales without requiring a priori knowledge of the task, making it a valuable contribution to ongoing work on model-based planning, representation learning, and lifelong learning algorithms.

In this paper, we study Reinforcement Learning from Demonstrations (RLfD) that improves the exploration efficiency of Reinforcement Learning (RL) by providing expert demonstrations. Most of existing RLfD methods require demonstrations to be perfect and sufficient, which yet is unrealistic to meet in practice. To work on imperfect demonstrations, we first define an imperfect expert setting for RLfD in a formal way, and then point out that previous methods suffer from two issues in terms of optimality and convergence, respectively. Upon the theoretical findings we have derived, we tackle these two issues by regarding the expert guidance as a soft constraint on regulating the policy exploration of the agent, which eventually leads to a constrained optimization problem. We further demonstrate that such problem is able to be addressed efficiently by performing a local linear search on its dual form. Considerable empirical evaluations on a comprehensive collection of benchmarks indicate our method attains consistent improvement over other RLfD counterparts.

"It is now too late to stop a future collapse of our societies because of climate change." These are not the words of a tinfoil hat-donning survivalist. This is from a paper delivered by a senior sustainability academic at a leading business school to the European Commission in Brussels, earlier this year. Before that, he delivered a similar message to a UN conference: "Climate change is now a planetary emergency posing an existential threat to humanity." In the age of climate chaos, the collapse of civilization has moved from being a fringe, taboo issue to a more mainstream concern. As the world reels under each new outbreak of crisis--record heatwaves across the Western hemisphere, devastating fires across the Amazon rainforest, the slow-moving Hurricane Dorian, severe ice melting at the poles--the question of how bad things might get, and how soon, has become increasingly urgent. The fear of collapse is evident in the framing of movements such as'Extinction Rebellion' and in resounding warnings that business-as-usual means heading toward an uninhabitable planet. But a growing number of experts not only point at the looming possibility that human civilization itself is at risk; some believe that the science shows it is already too late to prevent collapse. The outcome of the debate on this is obviously critical: it throws light on whether and how societies should adjust to this uncertain landscape. Yet this is not just a scientific debate. It also raises difficult moral questions about what kind of action is warranted to prepare for, or attempt to avoid, the worst. Scientists may disagree about the timeline of collapse, but many argue that this is entirely beside the point. While scientists and politicians quibble over timelines and half measures, or how bad it'll all be, we are losing precious time.

An estimated 165.8 million people shopped between Thanksgiving Day and Cyber Monday in 2018, and each purchased an average of 16 gifts during that time frame. Collectively, their haul came in billions of packages delivered through the mail system, over 900 million of which were tendered by the USPS alone. So what's a shopper stuck juggling multiple orders from multiple retailers to do? One solution is Route's tracking platform, which offers solutions on both the merchant and consumer sides. The Silicon Slopes, Utah-based startup today announced the launch of a visual order tracking app -- Route App -- for iOS, alongside a package insurance coverage plan, coinciding with the closure of $12 million in seed funding.

Google has announced that it's ending a controversial program that targeted homeless black people and scanned their faces to create AI training data in exchange for measly $5 gift cards. After an internal investigation prompted by news reports about the practice, Google says it's no longer sending third-party contractors out to gather face scans, according to New York Daily News. Instead, it will gather faces exclusively within Google campuses. But, bizarrely, a Google spokesperson is still defending the now-cancelled program. The original goal was to acquire face scans of dark-skinned people so that Google could present its facial recognition algorithms with a more diverse set of training data, a longstanding problem within AI that has led to widespread algorithmic bias against black people and other racial minorities.