• Journal of Nutrition and Health
• /
• v.33 no.1
• /
• pp.86-101
• /
• 2000

Various kinds of living circumstances are making population structure of Korean changed. That is, number of the children is decreased and that of the aged is increased. It is predicted that population of the children and the aged will be almost same until 2020. With above, as the expectation of the aged on healthy living might be increased, some special programs for the aged will be needed strongly. At this point, Korean aged population might be economically poor, comparing with other generation. In general, economic factor affects the subjective living-satisfaction and health status of them. Moreover, educational status, household shape and family tieing also affect their health status. According to the foreign articles, health status of the aged might be related to income, educational status, job, employed or/not, marital status, family structure, sex, and childhood condition. decrease of the income or unemployment could make the death rate of the aged higher. During childhood, discordance among the family might affect their health status after. IGUR is also important factor to affect the adulthood health. Positive life style of the aged would lessen their unequality of the health among them. Nutritional status of the Korean aged population might be indicated under the nutritional recommendation. It is affected by their income, education level, social class, and residing place. (Korean J Nutrition 33(1) : 86-101, 2000)

• Journal of Korean Academy of Nursing
• /
• v.18 no.2
• /
• pp.162-188
• /
• 1988

The purpose of this study is to guide the direction of the Korean nursing education to analysize ⑴ the philosophy and objectives ⑵ curriculum, and ⑶ educational environment. This analysis is based on the data from 50 nursing schools (14 4-year colleges and 35 3-year colleges) The survey was conducted from Dec. 1986 through Jan. 1987 by mail. 1) Educational philosophy and objectives 10 4-year colleges and 8 3-year college program have curricular philosoph. Most popular curricular philosophies are human beings, health, nursing, nursology, nursing education, nurses role in the present and in the future. 10 nursing schools mentioned that human being is the subject to interact with : environment physically, mentally and socially. 2 schools mentioned that health is the state of functioning well physically, mentally and socially. 13 schools mentioned that the nursing is the dynamic act to maintain and to promote the highest possible level of health. 4 schools mentioned that the nursology is an applied science. 4 schools mentioned that nursing education is the process to induce the behavioural changes based on the individual ability. There is different opinion about the nurses' role between 4-year college and 3-year college. In the responses from 4-year colleges they focus on the leadership in effective changes, self-regulating and self-determining responsibilities, applying the new technology, continuing education, and participation in research to further nursing knowledge. In the responses from 3-year colleges, they focus on the education in college, primary health care nursing, direct care provider and public health education. Among 50 respondents 40 schools have educational goals which can be divided into two categories. One is to establish the moral and the other is to develop the professionalism. 2) Curriculm The analsis of curriculum is only based on the data from the 4-year colleges because the most of 3-year colleges follow the curriculum guideline set by the Ministry of Education. a) Comparison of the credits in cultural subject and in nursing major. The average required credit for graduation is 154.6 and the median credit is the range of 140-149. The average credit of cultural subjects is 43.4. In detail, the average number of credit of required course and elective courses are 24.1 and 19.3 respectively. The average credit for major subject is 111.2. In detail, the average credit for required courses and electives course are 100.9 and 10.4 respectively. In 5 colleges, students are offered even on elective course b) Comparison of the credit by class. The average earned credits are as follows : 41.1 in freshman, 400 in sophormore 38.3 in junior and 32.4 in senior. Cultural subjects are studied in early phases. c) Comparison of the compulsory and elective cultural subject by institute. The range of credit is 7-43 in compulsory cultural subjects and there are lot of differences among institutions. While all respondents require liberal arts as compulsary subjects, few respondents lists social science, natural science and behavioral science as required subjects. Social science-related subjects are frequently chosen as cultural subjects d) Distribution of creditsin cultural subjects by institute. The liberal art subjects are taught in 20 institute. English and physical education courses are taught in all instituions. The social science subjects are taught in 15 colleges and the basic Psycology and the Basic sociology are the most popular subjects. The natural science subjects are taught in 7 colleges and Biology and Chemistry are the most popular subjects among them. e) Distribution of credits in major basic courses by institute. Most of the institutes select Anatomy, Microbiology, Physiology, biochemistry and Pathology as basic major courses. f) Comparison of the required and elective courses for nursing major by institutions. Subjects and credit ranges in major are varing by institute. More than half of the respondents select the following subjects as required major subjects. (1) Adults Health Nursing and Practice (19.5 credits) (2) Mother and Child Care and Practice (8.9 credits) (3) Community Health Care and Practice (8.5 credits) (4) Psychiatric Nursing Care and Practice (8.1 credits) (5) Nursing Management and Practice (3.9 credits) (6) Fundamental of Nursing, Nursing Research and Health Assessment and Practice. Three institutions select Introduction to nursing, Rehabilitation Nursing, School Nursing, Public Health Nursing, Nursing English, Communication, Human Development as electives in nursing major. 3) Educational environment a) Nursing institution There are forty-three 3-year colleges and seventeen 4-year colleges and 81.4% of which are private b) Number of students and faculty 19.2% of the students are in 4-year colleges and 80.8% of the students are in 3-year colleges. In 4-year colleges, the number of nursing faculty members is in the other of assistant professor, instructor and professor. In 3-year colleges, the orderiis lecturer, associate professor, full time instructor and assistant professor. In 4-year colleges, 18.8 students are allocated per nursing faculty and in 3-year colleges, 33.1 students are allocated per nursing faculty. c) Clinical practices 66.7% of the 4-year colleges practice over 1201 hours in clinic and 28.5% of 3-year colleges practice over 1201 hours in clinic. In 4-year colleges, 11.5 students are allocated per nursing faculty and in 3-year colleges,17 students are allocated per nursing faculty The survey shows no difference in the procedure between 4-year colleges and 3-year colleges but 3-year colleges choose the more variety practicing site such as special hospital and community health clinic. d) Audiovisual facilities The survey shows a lot of difference in audiovisual facilities among institution and 3-year colleges are less equipped than 4-year colleges.

• The Korean Nurse
• /
• v.36 no.1
• /
• pp.108-117
• /
• 1997

This study was carried out to examine the standards for evaluation of laboratory facilities and equipment. These constitute the most important yet vulnerable area of our system of higher education among the six school evaluation categories provided by the Korean Council for University Education. To obtain data on the present situation of holdings and management of laboratory facilities and equipment at nursing schools in Korea, questionnaires were prepared by members of a special committee of the Korea Nursing Education Society on the basis of the Standards for University Laboratory Facilities and Equipment issued by the Ministry of Education. The questionnaires were sent to nursing schools across the nation by mail on October 4, 1995. 39 institutions completed and returned the questionnaires by mail by December 31 of the same year. The results of the analysis of the survey were as follows: 1. The Physical Environment of Laboratories According to the results of investigation of 14 nursing departments at four-year colleges, laboratories vary in size ranging from 24 to 274.91 pyeong ($1{\;}pyeong{\;}={\;}3.3m^2).$. The average number of students in a laboratory class was 46.93 at four-year colleges, while the number ranged from 40 to 240 in junior colleges. The average floor space of laboratories at junior colleges, however, was almost the same as those, of laboratories at four-year colleges. 2. The Actual State of Laboratory Facilities and Equipment Laboratory equipment possessed by nursing schools at colleges and universities showed a very wide distribution by type, but most of it does not meet government standards according to applicable regulations while some types of equipment are in excess supply. The same is true of junior colleges. where laboratory equipment should meet a different set of government standards specifically established for junior colleges. Closer investigation is called for with regard to those types of equipment which are in short supply in more than 80 percent of colleges and universities. As for the types of equipment in excess supply, investigation should be carried out to determine whether they are really needed in large quantities or should be installed. In many cases, it would appear that unnecessary equipment is procured, even if it is already obsolete, merely for the sake of holding a seemingly impressive armamentarium. 3. Basic Science Laboratory Equipment Among the 39 institutions, five four-year colleges were found to possess equipment for basic science. Only one type of essential equipment, tele-thermometers, and only two types of recommended equipment, rotators and dip chambers, were installed in sufficient numbers to meet the standards. All junior colleges failed to meet the standards in all of equipment categories. Overall, nursing schools at all of the various institutions were found to be below per in terms of laboratory equipment. 4. Required Equipment In response to the question concerning which type of equipment was most needed and not currently in possession, cardiopulmonary resuscitation (CPR) machines and electrocardiogram (ECG) monitors topped the list with four respondents each, followed by measuring equipment. 5. Management of Laboratory Equipment According to the survey, the professors in charge of clinical training and teaching assistants are responsible for management of the laboratory at nursing schools at all colleges and universities, whereas the chief of the general affairs section or chairman of the nursing department manages the laboratory at junior colleges. This suggests that the administrative systems are more or less different. According to the above results, laboratory training could be defined as a process by which nursing students pick up many of the nursing skills necessary to become fully qualified nurses. Laboratory training should therefore be carefully planned to provide students with high levels of hands-on experience so that they can effectively handle problems and emergencies in actual situations. All nursing students should therefore be thoroughly drilled and given as much on-the-job experience as possible. In this regard, there is clearly a need to update the equipment criteria as demanded by society's present situation rather than just filling laboratory equipment quotas according to the current criteria.

• Journal of Gifted/Talented Education
• /
• v.14 no.2
• /
• pp.1-18
• /
• 2004

This study attempted to survey the present status of the eight students who, 17 years from now, had participated in a special program for 170 days preparing for early entrance to the KAIST. With six students who responded to the questionnaire previously sent, the study focused on each participant's achievement at present, his evaluation on Accelerated Education, his personal opinions on early entrance to university and Gifted Science Education. From the survey, we could find out that three participants had received a doctorate and they are currently working as researchers at research center or professors at university. The other three participants are working as researchers at research center after receiving the M. S. Academic achievements vary according to each participant, but three of them, who ranked high in research achievement, had their articles published in international journals of the SCI level even when they were between 25 and 28 of age. Two of them in particular showed outstanding engineering and developing the commercial modem for IMT 2000 for the first time in Korea. One participant entered Harvard University to major in physics and afterwards studied the law in UCLA. He is now a professor in the law department. Concerning the evaluations on Accelerated Education and early entrance system, four participants answered positively while two participants expressed negative attitudes. When asked their opinions on the Gifted Education in Korea, such problems were pointed out as too much pressure of entering the first-class universities, the tendency to avoid choosing science-related major at universities, the lack of education system that encourages creative thinking, etc. It is necessary that the accelerated education play an important role to maximize their hidden potentials. It remains to be seen in what places these six participants will find themselves and of what ideas they will come up within the next 17 years. It will be worthwhile to wait until then for another survey.

• Journal of Gifted/Talented Education
• /
• v.14 no.4
• /
• pp.125-149
• /
• 2004

The purpose of this study is to provide basic information on the current status of elementary school teachers' perception of gifted education. For this purpose, this study wi1l analyze elementary school teachers' perception of the gifted education (i.e., general perception of gifted education, characteristics of gifted children, identification of gifted children, programs for gifted education, and teachers of gifted education). A questionnaire survey was used for the purpose and research questions of this study. The questionnaire used in this study was constructed by taking into account the results of surveys conducted in previous studies and the literature on gifted education. Before conducting the research, a preliminary inquiry was made to identify problems that may occur while the subjects were participating in the survey, as well as to determine the appropriateness of the questionnaire and the amount of time needed. The preliminary inquiry was conducted with ten randomly selected elementary school teachers who did not participate as subjects in the actual research. The results were later used as initial data for the actual research. The subjects of this study were teachers who were teaching in8 elementary schools under each office of education in Seoul. This process was conducted for 180 elementary school teachers from April to May 2004. The results were analyzed using SPSS (Statistical Package for Social Science) Ver. 10.1, a software program for statistical research. After the data were analyzed, the following conclusions were arrived at: 1. The result of the genera1 perception of gifted education by elementary school teachers were positive and reasonably high. The level of their perception of detailed information or knowledge, however, was relatively low. 2. As for their perceptions of the emotional characteristics of gifted children, t11e results showed a low level of understanding of the characteristics of gifted children. 3. As for their perceptions of identification of gifted children, the results showed a high level of understanding of the appropriate time to provide special education to gifted children and of the methods to identify such. On the other hand, their understanding of the identification of gifted children in an actual class was poor. 4. The respondents' level of perceptions of programs for gifted education was very low since many subjects did not have any experiences with such programs. 5. The results showed a very positive response to receiving training on gifted education, though they were very reluctant to be assigned as teachers of gifted education because of the excessive work that is associated with such and their lack of capability in handling gifted children.

• Education of Primary School Mathematics
• /
• v.16 no.2
• /
• pp.123-145
• /
• 2013

This study was expected to yield the meaningful conclusions from the experimental group who took lessons based on inductive activities using GeoGebra at the beginning of proof learning and the comparison one who took traditional expository lessons based on deductive activities. The purpose of this study is to give some helpful suggestions for teaching proof to mathematically gifted elementary students. To attain the purpose, two research questions are established as follows. 1. Is there a significant difference in proof abilities between the experimental group who took inductive lessons using GeoGebra and comparison one who took traditional expository lessons? 2. Is there a significant difference in proof attitudes between the experimental group who took inductive lessons using GeoGebra and comparison one who took traditional expository lessons? To solve the above two research questions, they were divided into two groups, an experimental group of 10 students and a comparison group of 10 students, considering the results of gift and aptitude test, and the computer literacy among 20 elementary students that took lessons at some education institute for the gifted students located in K province after being selected in the mathematics. Special lesson based on the researcher's own lesson plan was treated to the experimental group while explanation-centered class based on the usual 8th grader's textbook was put into the comparison one. Four kinds of tests were used such as previous proof ability test, previous proof attitude test, subsequent proof ability test, and subsequent proof attitude test. One questionnaire survey was used only for experimental group. In the case of attitude toward proof test, the score of questions was calculated by 5-point Likert scale, and in the case of proof ability test was calculated by proper rating standard. The analysis of materials were performed with t-test using the SPSS V.18 statistical program. The following results have been drawn. First, experimental group who took proof lessons of inductive activities using GeoGebra as precedent activity before proving had better achievement in proof ability than the comparison group who took traditional proof lessons. Second, experimental group who took proof lessons of inductive activities using GeoGebra as precedent activity before proving had better achievement in the belief and attitude toward proof than the comparison group who took traditional proof lessons. Third, the survey about 'the effect of inductive activities using GeoGebra on the proof' shows that 100% of the students said that the activities were helpful for proof learning and that 60% of the reasons were 'because GeoGebra can help verify processes visually'. That means it gives positive effects on proof learning that students research constant character and make proposition by themselves justifying assumption and conclusion by changing figures through the function of estimation and drag in investigative software GeoGebra. In conclusion, this study may provide helpful suggestions in improving geometry education, through leading students to learn positive and active proof, connecting the learning processes such as induction based on activity using GeoGebra, simple deduction from induction(i.e. creating a proposition to distinguish between assumptions and conclusions), and formal deduction(i.e. proving).

• Journal of Korean Home Economics Education Association
• /
• v.20 no.1
• /
• pp.45-62
• /
• 2008

The purpose of this study was to investigate middle school home economics(HE) teachers' perception and needs on self supervision related to HE subject matter, Using the methods of survey and interview, 177 samples were collected. For collected surveys, mean value, standard deviation, frequency, percentage analysis were performed by using an SPSS/Win (ver10.1) program. The results of this study were as follows. First, the middle school HE teachers recognized that self supervision related to HE subject matter was absolutely needed to expand the improvement of techniques for teaching instructions and the width of knowledge on the studies on textbook. Second, the middle school HE teachers recognized the necessary parts of self supervision related to HE subject matter as HE teaching-learning methods, the studies on textbook contents, and HE education philosophy in order. Third, the middle school HE teachers recognized that it would be helpful in improving their HE class and expertise in order of field survey, participation in various training programs, utilization of mass media, participation in societies for researches and meetings and information sharing with co-teachers among the types of self supervision. Fourth, the middle school HE teachers needed the reduction in miscellaneous duties, less pressure for time, restoration of teachers' desire, support of physical resources (improvement of various environments such as classrooms and special rooms), economic support and various support programs (expanding the opportunities to participate in training and society and establishment of a database for relevant materials, etc.) to facilitate self supervision. As such, the middle school HE teachers' overall recognition on HE-related self supervision became significantly higher. To enhance the HE-related expertise, however, it would be necessary to conduct concrete and active support for HE education, philosophical area and the studies on textbook contents as well as the teaching-learning methods for HE in which teachers' demand was high. In addition, the HE teachers wanted to have an easy and quick access to various HE-related data; therefore, it would be urgent to summarize scattered relevant data and support the HE teachers more systematically.

• Journal of Elementary Mathematics Education in Korea
• /
• v.10 no.1
• /
• pp.67-87
• /
• 2006

This study is to recognize a problem in the practice of performance assessment in elementary school, and to find out some suggestive points for improvement of teaching·learning method in elementary mathematics through assessment by reducing time restriction according to assessment through the practice of self-assessment, peer-assessment and observation, and then by reflecting the results of assessment on teaching learning plan. For that, the questions of study set up are as follows ; 1. How should self-assessment and peer-assessment be applied to in elementary mathematics assessment? 2. How should the time for 'let's play an interesting game' be managed for assessment of elementary mathematics? 3. How should the results of assessment be reflected on the Process of teaching and learning of mathematics? To solve these problems, a researcher of this thesis performed self-assessment, peer-assessment on 40 students of second grade under her charge as a class teacher, and applied observation in the time management process for 'let's play an interesting game' for a semester. self-assessment was made by mathematics journal, self-assessment chart, peer-assessment was by the analysis of conversation record among students in the situations of assessment, and observation was by observation of activity when playing with the results data of play analyzed. the concrete methods of application as follows ; First, mathematics journal was applied $1{\sim}2$ times by each unit with reconstruction into the level of second grade on the basis of the preceding-study models. Second, peer-assessment was applied to the unit-assessment time and the play-activities time by the method of recording·analyzing the contents of conversation among students in the process of assessment. Third, mathematical attitude & dispositions of students making use of the self- assessment table were examined referring to the teaching learning plan. Fourth, the time management for 'let's play an interesting game' was made through the prior recognition of play method and the joyful play-activities by use of the play-plate. Assessment depended on analysis of play-activities results of students making use of an observation form. Fifth, the results of self-assessment, peer-assessment, and observation were analyzed, and then they were made use of as self-observation data, of teacher her/his self, or teaching·learning improvement data. Students' self-assessment datum (mathematics diary, self-assessment sheets, conversation contents in the process of assessment) and observation materials (check lists, Play-activity result materials, conversation contents in the process of play) obtained in the process of application was analyzed as follows ; 1. From the practice of self-assessment in form of mathematics journal, I could obtain not only datum showing how much students was understanding the learning aims by unit time and to any degree they reached but also information about their response to learning datum and favorable type of learning. 2. Assessment by self-assessment chart was useful in planning the mathematics teaching learning process because it helps ascertain mathematical attitude & dispositions of students. 3. Through the application of peer-assessment, students had the opportunity of communicating with other students looking back on his/her explaining process, and teachers could obtain basic materials for assessment of students. 4, In case of time management for 'let's play an interesting game', there was natural extension of play made through time-security by prior looking into the method of play-activity, and then, for a remained time, by making children play a new game. 5, I could easily record the activities of students by use of the observation. form, and make use of it as basic data for descriptive assessment. 6, Each kinds of data obtained from the results of assessment was helpful for securing self-observation materials in the process of teaching learning and for their betterment in mathematics subject. However, because they were in the second grade of elementary school and there was an individual difference, some students could not make use of mathematics diary or self-assessment form properly. In case of these students, assessment data would be obtained through interview or observation. And for effective operation of play, its purpose & method and matters that demand special attention when play-acting should be clearly guided. Also, when applying an effective play in addition to play activities in textbook, to lessons, interesting mathematics lessons could be guided.

• Journal of Elementary Mathematics Education in Korea
• /
• v.13 no.2
• /
• pp.163-192
• /
• 2009

Currently, our country operates gifted education only as a special curriculum, which results in many problems, e.g., there are few beneficiaries of gifted education, considerable time and effort are required to gifted students, and gifted students' educational needs are ignored during the operation of regular curriculum. In order to solve these problems, the present study formulates the following research questions, finding it advisable to conduct gifted education in elementary regular classrooms within the scope of the regular curriculum. A. To devise a teaching plan for the gifted students on mathematics in the elementary school regular classroom. B. To develop a learning program for the gifted students in the elementary school regular classroom. C. To apply an in-depth learning program to gifted students in mathematics and analyze the effectiveness of the program. In order to answer these questions, a teaching plan was provided for the gifted students in mathematics using a differentiating instruction type. This type was developed by researching literature reviews. Primarily, those on characteristics of gifted students in mathematics and teaching-learning models for gifted education. In order to instruct the gifted students on mathematics in the regular classrooms, an in-depth learning program was developed. The gifted students were selected through teachers' recommendation and an advanced placement test. Furthermore, the effectiveness of the gifted education in mathematics and the possibility of the differentiating teaching type in the regular classrooms were determined. The analysis was applied through an in-depth learning program of selected gifted students in mathematics. To this end, an in-depth learning program developed in the present study was applied to 6 gifted students in mathematics in one first grade class of D Elementary School located in Nowon-gu, Seoul through a 10-period instruction. Thereafter, learning outputs, math diaries, teacher's checklist, interviews, video tape recordings the instruction were collected and analyzed. Based on instruction research and data analysis stated above, the following results were obtained. First, it was possible to implement the gifted education in mathematics using a differentiating instruction type in the regular classrooms, without incurring any significant difficulty to the teachers, the gifted students, and the non-gifted students. Specifically, this instruction was effective for the gifted students in mathematics. Since the gifted students have self-directed learning capability, the teacher can teach lessons to the gifted students individually or in a group, while teaching lessons to the non-gifted students. The teacher can take time to check the learning state of the gifted students and advise them, while the non-gifted students are solving their problems. Second, an in-depth learning program connected with the regular curriculum, was developed for the gifted students, and greatly effective to their development of mathematical thinking skills and creativity. The in-depth learning program held the interest of the gifted students and stimulated their mathematical thinking. It led to the creative learning results, and positively changed their attitude toward mathematics. Third, the gifted students with the most favorable results who took both teacher's recommendation and advanced placement test were more self-directed capable and task committed. They also showed favorable results of the in-depth learning program. Based on the foregoing study results, the conclusions are as follows: First, gifted education using a differentiating instruction type can be conducted for gifted students on mathematics in the elementary regular classrooms. This type of instruction conforms to the characteristics of the gifted students in mathematics and is greatly effective. Since the gifted students in mathematics have self-directed learning capabilities and task-commitment, their mathematical thinking skills and creativity were enhanced during individual exploration and learning through an in-depth learning program in a differentiating instruction. Second, when a differentiating instruction type is implemented, beneficiaries of gifted education will be enhanced. Gifted students and their parents' satisfaction with what their children are learning at school will increase. Teachers will have a better understanding of gifted education. Third, an in-depth learning program for gifted students on mathematics in the regular classrooms, should conform with an instructing and learning model for gifted education. This program should include various and creative contents by deepening the regular curriculum. Fourth, if an in-depth learning program is applied to the gifted students on mathematics in the regular classrooms, it can enhance their gifted abilities, change their attitude toward mathematics positively, and increase their creativity.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.