In the era of artificial int≠≥elligence, intelligent assistant must t γ"σake the master as the center&quo♦©±t;

    NetEase techn  ology news July 9th ne€↑ ws, according to foreign media repo♠♦rts, 2016 is known as the first year of the roboΩΩt chat. With the development of conveδ' ∑rsation technology, cha ™t robot will develop in>≥←to intelligent assistant. The intelliδ"‌εgent assistant needs to ru¶★÷₩n in the framework o'✔$λf the determination, li∑÷>ke the three law of '✔φthe robot to follow the prin ¶ciple of determination.

    Apple to open Siri inte♦ "rface to the developer, Messenger Fac£♣ebook also launched a chat robot.

    Althou•©≥•gh some aspects of the chat↓★&> robot's function is diγ← ★sappointing, but it is indeed ♣β‌an important step in the develop±©↓ment of conversation techno>¶logy. Dialogue technology will go to w✔​‍♥here? The author believes that>₩∞ we are in the forefront of≥ <≠ intelligent technology. The deve•&£lopment of technology such as Natural L×δ↔anguage Processing, ar←∑tificial intelligence and conversΩ± ©ational interface makes it possibl ∞e to realize the intelligent assistant.

    With the growing •>÷development of chat robot into a real intellig'•ent assistant, which can help users do a lo✔§  t of things, and human autonomy ≥$ ∏is also facing many challenges. ®₩©®An intelligent assistant needs to run under a cer&™®→tain framework, including how'‍₽‍ to interact with the human, how to make a&±λ∏ decision, how to understand and use the informat <↕↓ion obtained.

    At present, we have developed®™ an intelligent assistant for the intelle↓£ctuals, and also follow the Asimov's three §¶law of the robot for intelligent digita φקl assistant to set up t ∞£hree principles, they are:

    1≤₩. intelligent assistants must a♥ ¥€lways serve the user

    If the pro↓¥blem of intelligent assist★♥↔✔ant than it can solve many,σδ it is useless for users. A qualified intellig✔∞​∞ent assistant should meet the n≥™↕≥eeds of users, can provide ✘ ✘σinformation and decision-maki☆↑λng reference. An intelligent assistant shou↑¥∑↕ld be able to perform a simple task for ☆↕↓≈a user. In addition, theα© work should be visualized to facilitate th£↔≥↔e user to understand the work of the int☆'₽elligent assistant and make related decisions.

    For example, i φ ♣n the early days of in↑ σtelligent assistants and us₹δφ∏ers, the corresponding artificial ¶•γ∞intelligence should be provided to the user ♥ ™to provide the implementat₽≈ion process and decision-mak$™×σing process transparency. A<"s a human assistant, once an arti♣®σficial intelligence in tφ♠×£he early work to establish α ×‍a class trust and demonstrat♣₽≤λe the ability, then it is a b™'↑₩ig job to report results after the completi≥™on of the work.

    This meanφק®s that the smart assistant•♥™ does not need to se←✔&≥t goals, it's only goal is to serve the user. Onl'↓♠y if the user is successful, intelligent as®↕'sistant to be successful, th¶ φis is the core of their ±‍←φwork.

    2 .intelligent assi<¥σ stant must represent the maximum benefit of the u≠δser

    An intellige>"nt assistant should be able to tak  €e the place of the user to carry ☆♠on the related trans‌✔ ¶action. It should understand the user'α✔✘s habits, preferences and pri←♣±orities in order to mee←≥>t the real needs of users. In some cases, usersλ' ask the work done by the '₹intelligent assistant does not have∑σ'€ the direction, such a™<$§s the arrangements for the meeting or to rent a c¥¶γar for me to go to New York". The intell ε ♦igent assistant only in-depth und★©βerstanding of the user, in order to♦'≥< make the most consistent $≈with the interests of users.

    In the future,∞♦×‍ we will be able to see the exc₩↔hange of two intelligent assistants. So££φ when the two artificial intelligence ↕™on behalf of the owners of their £ own willingness to communicate< →, you have to make sure that their own assistants§φ to meet their own interes÷¥ts. Part of the reason is ≥¶that your intelligen↕>✔×t assistant to help you rent a car rent↕₽★≤al company than the total£> number of intelligent assistant.

    3. intelli±>εgent assistants must l≥πearn to adapt and learn✘≠

    At first, an inte§♥lligent assistant should have a ∞$specific skill or function. But ove₹≥♠∞r time, it should be able to lea↓§♣rn, can perform more complex tas'∞®®ks.

    For the development of an in≤"€telligent assistant, this is the most <φ₹ difficult, because the ad<≠ aptation and learning and the way we pr>≤φ≥esent the software architecture. And the¶♥©÷ learning function of the software mean≤<s that we don't know how it will develop. Lea∑♠rning ability means that intelligent assista∞λαnts will be taught, then how can we preven∑ ±t another Tay Microsoft from appea♠↓±εring?

    For intelligent assistan ♠←ts can learn what and how to learn, you need to≠¥∞ have the relevant rules. This will be ab¥£le to ensure that it will not be c∞₩ ontrary to the wishes of tπ©he user.

In the next five yearsβ≠•, intelligent assistants will chang✘γe the way we work. Every day, we will use intellλ€igent assistants to carry out simple work, we wil≥ l be released from a number o π f tedious work, let us have more time and expe&×rience in the work of love. (Ni ∞ng Yu)