These two volumes are the papers presented at the International Conference on Neural Information Processing (ICONIP'96) held in Hong Kong, 24-27 September 1996. Over three hundred selected papers make up two volumes of 700 pages each. Featured topics of Volume 1 are Learning Theories and Algorithms, Speech and Signal Processing, Pattern Recognition and Image Processing, Recurrent Networks, Automata and Dynamics. Volume 2 focusses on Financial Engineering and Time Series Forecasting, Evolutionary Computing, Neurosciences and Biophysics, Neural Control and Robotics, Hardware Implementations, Hybrid Systems and Applications, plus a number of Special Sessions. Other topics covered include image processing, computer vision, fuzzy logic, pattern recognition, algorithms and architectures, and financial engineering. These two volumes combine to cover the major topics in neural information processing and reflect the latest progress with a good balance between scientific studies and industrial applications, a well as featuring neural information processing approaches on financial engineering.