Conversation is unequivocally a human capability, differentiating us from the rest of creation, forming the basis of culture and civilization, and defining our unique level of intelligence as 'species. It serves many purposes in our daily lives: communication, coordination, social bonding, completing complex tasks, education, comfort, and entertainment, to name a few. There's nothing like a good lively dialogue to get us excited and creative about a subject or... about each other. A quick retort makes us laugh or persuades us who to choose for our next president! Humans are really good at striking up conversations, whether it's for work or just shooting the breeze. When we speak to ourselves, we are constantly leveraging information and contextual knowledge to convey sarcasm, read between the lines, and express our personality. Conversation is a high bandwidth channel where knowledge, instructions, behaviors, emotions, force of will and many other messages are transmitted via language (written or spoken) and its structure. In our daily lives, we usually engage in oral conversations, while recent technological innovations have introduced us to the habit of chatting in an almost instantaneous, typed manner. Say no to plagiarism. Get a tailor-made essay on “Why Violent Video Games Should Not Be Banned”? Get an original essayThe complexity of the presentation, the structure, the conduct of the interaction and the amount of information it conveys are all measures of the intelligence of the participants. . We are surrounded by an animal world of inarticulate cries transmitting simple messages, and a cold world of systems and machines that require specialized and limited forms of instruction and manipulation. Thus, humanity has dreamed of naturally interacting with tools that will carry out its orders. Popularized in science fiction novels, the concept has existed since the time of Homer (see Ulysses Rhapsody Σ where Vulcan is served by golden maids resembling humans). AI arrives thousands of years later with the promise that it will make humanity's dream come true: having "intelligent" conversations with our machines, in the sense that we will be able to obtain information, transmit instructions , gain education or even get advice in a natural way. In fact, the measure of success of these conversational AI systems is none other than the Turing test, which is also a measure of intelligence: having a conversation with a machine about a subject or task that would be impossible to distinguish from speaking to another human. However, besides human narcissism and creativity, what is further fueling this push for intelligent conversational systems and assistants is a long list of enterprise applications and a strong market need for personalized conversations between businesses and their customers. Until now, the complexity and limitations that existing dialog tools and resulting conversational systems suffer from have served businesses and their customers with disappointing experiences. They eventually get the job done, but after great effort, high development and maintenance costs, and relatively limited human interaction experience. Indeed, today's automated conversational systems are not really intelligent! Designers with domain knowledge and IT expertise define and program each conversation with the scripted responses users can expect when interacting withautomated dialogue systems. Thus, conversational systems are built based on decision tree logic, where the response given by the bot depends on a dialog state defined by specific intentions and keywords identified in the bot's input. user. IF user input contains “shop” or “purchase” (intent); AND “cell phone” or “mobile” (product type); THEN send a message with the cell phone list. Generally, designers need to program 3 main components in order to create an automated conversational system: a) its natural language understanding part, i.e. the part that parses and analyzes human language and identifies the parts that are important for the task, b) the dialog management part, which essentially identifies a state of the dialog based on history and current parsed input in order to decide what to do next, and c) a response generation part, where typically the designer programs the system's scripted responses. All this means that the resulting systems will seem as intelligent as the effort (and patience) put in by the designers who created them: capturing and anticipating large numbers of potential use cases and inputs, creating appropriate responses and natural. Moreover, adapting and maintaining such rule-based dialog systems with changing or new information about a task is a labor-intensive and time-intensive programming job. To mitigate these shortcomings, next-generation dialog systems must be capable of learning. To begin with, there are two easily accessible sources of knowledge: a) examples of human-to-human interactions and b) existing data (books, websites, manuals, databases). After all, this is what businesses also have at their fingertips. Companies have collected enormous amounts of conversation examples from their agents' interaction with customers over the phone or other channels (online, Twitter, etc.). Similarly, businesses have abundant organized (databases, knowledge graphs, websites) or raw (documents, manuals) data that contains knowledge related to business operations and a variety of business services (travel booking) or targeted tasks (maintenance, repair). However, so far, several attempts at automated systems based on examples have not been successful and have resulted in unexpected, even comical results (remember the Microsoft Tay offensive and the amusing Facebook trading bot challenge ). the data you feed into the system when you train it and also the mechanism you use to produce answers. Such systems will only be accepted in the business world when their responses are not free-form, but can be limited into a set of responses acceptable to the business. The above highlights a deeper truth: today's automated conversational systems do not make the connection with available knowledge. This connection and transfer of knowledge from data has so far been carried out directly by human design and programming. So, when creating an automated dialog system, we need to recreate existing functionality from scratch, i.e. recreate our website experience in a different way, while the underlying information underlying are the same. This means extra work for businesses to create an entirely different channel to handle customer requests, which also results in a lack of consistency for the user. On the other hand, human agents have nonot this problem. They are able to continually acquire knowledge by observing others or from documents, assimilate new information, and organize themselves appropriately in ways that increase their ability to lead new goal-directed conversations. Transferring this human capacity, even to a certain extent, to our automated dialogue agents and thus learning from dialogue examples and existing knowledge sources, is necessary for progress and it is clear that the market wants it. If we can appropriately represent knowledge about a goal-directed task and automatically integrate it into a neural network or traditionally programmed dialogue systems, we hope to have several advantages. First of all, we hope to improve the performance/response accuracy of the systems. No longer depend on the degree of rigor and expertise of the system designer; the system will learn everything there is to know from the available data. More and more, we will be able to infuse a little “common sense” into systems that can be transmitted from one application to another. Systems will be able to quickly adapt to new and changing information as well as operate and evolve in the absence of dialogue examples. Yet this link between dialogue and knowledge remains elusive – it is a difficult problem for machines. Leveraging knowledge and restricting conversational systems to a set of acceptable responses will greatly automate their development and maintenance, but that doesn't mean we don't still need to spend effort programming them or improving their comprehension capabilities of language. Conversational interfaces represent a big change in the way we are used to thinking about interacting with our “dumb” computers and “smart” phones! Conversational computing is a paradigm shift that requires designers to change their thinking, deliverables, and design process in order to create successful bot experiences. We therefore hope that major progress will also be made in the systems and tools allowing the composition and integration of the different components linked to the dialogue. Designers must be able to take advantage of the strengths and functionalities offered by different AI technologies, either to analyze human language or to conduct dialogues drawn from different sources. So, over the next few years we will see a proliferation in the market of tools that will not only support the creation of conversational systems from scratch, but also allow scaling up to similar tasks, transferring knowledge and manage dialogue, and continually incorporate new data. Such conversational interfaces will become the new operating system or digital mesh that will hold technologies together. Our future will be flooded with digital assistants, drones, robots and self-driving cars. Therefore, we must also look for innovative ways to converse with these new devices. This means not only giving one-way instructions or requests, but also conducting two-way interactions that meet our needs. This is where conversational computing comes in. We need conversation not just to fill out forms or for step-by-step instructions, but we need it because we don't know the ever-changing options (e.g. tickets and dates available, or new situations encountered ) and the systems do notdo not know our needs, our preferences, or do not have our special training and wisdom at a given time to accomplish a task. Large companies are investing heavily in the conversational domain: Google, Apple, Amazon, Facebook, IBM, Baidu, to name just a few. And by mastering conversation, they can master the world. The next Alexa will be your home assistant or hotel concierge. The next Siri or Google Assistant will be your personal assistant in the office and at home. Facebook will interact with you as one of your friends. But beyond their dominance in our personal daily lives, conversational systems will take over the business world because they will provide faster, better and cheaper customer service. This is where companies like IBM and many start-ups are entering. According to Gartner, AI will account for 85% of customer relationships by 2020, and recent market analysis indicates that today, 60% of repeat customers (that's you and me) would prefer to speak to an automated system rather than a human. simple tasks, if they are faster and more informative. However, the majority of us (over 70%) still don't trust automated systems tasked with complex tasks or our money. Plus, according to the survey, most of us don't want to rely on automated assistants making decisions for us. These cases always require a human touch, someone who understands our needs, can negotiate, is able to explain and lead the conversation towards a win-win solution. Major progress is therefore expected in systems that demonstrate more human characteristics and that take into account more modes of interaction than simple typed messages. Microsoft, for example, is working on a natural user interface (NUI) that combines natural language with gestures, touch and gaze, to help deepen system conversations. Everything can be “felt” through sight, touch or hearing. This is the kind of multimodal conversation that will automate more human-like conversational systems. Google recently introduced Duplex, a concept assistant that makes appointments and reservations for you: it looks and interacts in a very human way. AI can play a major role in this. Research is already prototyping deep learning for increasingly “deeper” conversational systems: instead of learning dialogue from textual dialogue examples, new AI systems are in the works that will learn directly from interactions spoken. MILA, the UN Research Dialogue Team. of Montreal, Facebook, Samsung, Microsoft and Google are already working in this direction. This is going to be very powerful. Remember from our younger years that spoken dialogue was our main and only skill for learning, playing and teaching others. We were able to negotiate with our parents before we could read and write, and we were able to describe to our peers how to play a game, or coordinate to play it, before going to school. Interaction through oral dialogue is a rich form of intelligent communication that becomes more and more sophisticated and complex over the years: it continually integrates what we learn from our interaction with others or from sources of knowledge (books, documents, articles, manuals, etc.). about to see such systems in the near future? “You must have patience, my young Padawan. “Our current experience is with concatenated systems, that is, systems that have a speech-to-text component that transcribes our speech which is then transmitted to the conversational system and the response is read back to us. Of.