We transform reinforcement learning (RL) into a form of supervised learning (SL) by turning traditional RL on its head, calling this Upside Down RL (UDRL). Standard RL predicts rewards, while UDRL instead uses rewards as task-defining inputs, together with representations of time horizons and other computable functions of historic and desired future data. UDRL learns to interpret these input observations as commands, mapping them to actions (or action probabilities) through SL on past (possibly accidental) experience. UDRL generalizes to achieve high rewards or other goals, through input commands such as: get lots of reward within at most so much time! A separate paper [61] on first experiments with UDRL shows that even a pilot version of UDRL can outperform traditional baseline algorithms on certain challenging RL problems. We also introduce a related simple but general approach for teaching a robot to imitate humans. First videotape humans imitating the robot's current behaviors, then let the robot learn through SL to map the videos (as input commands) to these behaviors, then let it generalize and imitate videos of humans executing previously unknown behavior. This Imitate-Imitator concept may actually explain why biological evolution has resulted in parents who imitate the babbling of their babies.

Search methods, such as Genetic Algorithms and Simulated Annealing are typically used to achieve the required scalability in challenging problems for which it is hard to find optimal, or even just "good enough' solutions. The majority of these methods involve steps where the state of the algorithm is modified in some way to escape a local optimum. The aim is to avoid premature convergence, which is when the search method converges (usually very early in the search) to a local optimum of poor quality [1, 2]. Previous research has shown that the performance of search strategies is affected by the structure of the fitness landscape [3, 4]. A fitness landscape is defined by three components: i) the search space, which is the set of all candidate solutions, ii) the fitness function, which assigns a fitness value to each solution, and the neighbourhood operator, which defines how solutions are connected, and as a results how the search strategy can traverse the landscape.

The recognition of emotion and dialogue acts enrich conversational analysis and help to build natural dialogue systems. Emotion makes us understand feelings and dialogue acts reflect the intentions and performative functions in the utterances. However, most of the textual and multi-modal conversational emotion datasets contain only emotion labels but not dialogue acts. To address this problem, we propose to use a pool of various recurrent neural models trained on a dialogue act corpus, with or without context. These neural models annotate the emotion corpus with dialogue act labels and an ensemble annotator extracts the final dialogue act label. We annotated two popular multi-modal emotion datasets: IEMOCAP and MELD. We analysed the co-occurrence of emotion and dialogue act labels and discovered specific relations. For example, Accept/Agree dialogue acts often occur with the Joy emotion, Apology with Sadness, and Thanking with Joy. We make the Emotional Dialogue Act (EDA) corpus publicly available to the research community for further study and analysis.

Videos showing autonomous or self-driving vehicles weaving in and out of crossroads at speed without colliding suggest this technology will solve traffic problems. You almost never see pedestrians or cyclists in these videos. The reality is that they don't fit. The vision of autonomous traffic is either of a large convoy of vehicles just a metre apart moving along road corridors at 100km/h, or of vehicles in an urban setting where their sensors are picking up every pedestrian movement and slowing or stopping. In the first case, the vehicles form an impenetrable barrier to pedestrians or cyclists (who, like on a freeway, will probably be banned).

Last week, the Australian state of New South Wales announced a plan to crack down on drivers using their phones on the road. The state's transport agency said it had integrated machine vision into roadside cameras to spot offenders. The AI automatically flags suspects, humans confirm what's going on, and a warning letter is sent out to the driver. "It's a system to change the culture," the assistant police commissioner of New South Wales, Michael Corboy, told Australian media, noting that police hoped the technology would cut fatalities on the road by a third over two years. It seems an admirable scheme, top to bottom.

More than half of companies in Australia (54%) hope to invest in inventory planning and logistics, compared to 39% of organisations elsewhere in the world. Meanwhile, 46% of Australian firms plan to expand their CRM, compared to 39% of companies in other countries. READ MORE: Half of existing jobs will be extinct by 2025 – is HR safe? While 45% of global business leaders want to invest in industrial automation, only 28% of those in Australia plan to do the same. Overall, international respondents claim AI would boost their company's productivity. However, only 21% of leaders in Australia predict it would result in a lower headcount in their industry.

Scientists have made artificial nerve cells, paving the way for new ways to repair the human body. The tiny "brain chips" behave like the real thing and could one day be used to treat diseases such as Alzheimer's. A team from the University of Bath used a combination of maths, computation and chip design to come up with a way to replicate in circuit form what nerve cells (neurons) do naturally. Neurons carry signals to and from the brain and the rest of the body. Scientists are interested in replicating them, because of the potential that offers in treating diseases such as Alzheimer's, where neurons degenerate or die.

Today Microsoft announced an update to the Seeing AI app that will include new language output options, including Dutch, French, German, Japanese, and Spanish. The iOS exclusive app was first released in 2017 as a free tool to help people with visual impairments navigate day-to-day life. It's built around a series of different channels, which users can select depending on their particular needs or circumstances. For the first time Microsoft's Seeing AI app will speak in languages other than English. Today's update enables audio output in Japanese, German, Spanish, Dutch, and French In one channel, the app will read out the text of any document the iPhone's or iPad's front facing camera is pointed at.

We introduce a novel keyword-aware influential community query KICQ that finds the most influential communities from an attributed graph, where an influential community is defined as a closely connected group of vertices having some dominance over other groups of vertices with the expertise (a set of keywords) matching with the query terms (words or phrases). We first design the KICQ that facilitates users to issue an influential CS query intuitively by using a set of query terms, and predicates (AND or OR). In this context, we propose a novel word-embedding based similarity model that enables semantic community search, which substantially alleviates the limitations of exact keyword based community search. Next, we propose a new influence measure for a community that considers both the cohesiveness and influence of the community and eliminates the need for specifying values of internal parameters of a network. Finally, we propose two efficient algorithms for searching influential communities in large attributed graphs. We present detailed experiments and a case study to demonstrate the effectiveness and efficiency of the proposed approaches.

Multi-language speech datasets are scarce and often have small sample sizes in the medical domain. Robust transfer of linguistic features across languages could improve rates of early diagnosis and therapy for speakers of low-resource languages when detecting health conditions from speech. We utilize out-of-domain, unpaired, single-speaker, healthy speech data for training multiple Optimal Transport (OT) domain adaptation systems. We learn mappings from other languages to English and detect aphasia from linguistic characteristics of speech, and show that OT domain adaptation improves aphasia detection over unilingual baselines for French (6% increased F1) and Mandarin (5% increased F1). Further, we show that adding aphasic data to the domain adaptation system significantly increases performance for both French and Mandarin, increasing the F1 scores further (10% and 8% increase in F1 scores for French and Mandarin, respectively, over unilingual baselines).