#### Filter Results:

- Full text PDF available (17)

#### Publication Year

1991

2016

- This year (0)
- Last 5 years (2)
- Last 10 years (9)

#### Publication Type

#### Co-author

#### Journals and Conferences

#### Key Phrases

Learn More

- Xin Wang, Arun K. Jagota, Fernanda Botelho, Max H. Garzon
- NIPS
- 1995

above gives c 2 = 2c 1 =(c 1 +(2?c 1)). Clearly, c 2 is proportional to the reciprocal of the ratio. Thus, when is small, can be taken larger than when is large. This may be used to evolve the network eeciently in the beginning and slow it down later, while ensuring that 2-cycles are never retrieved. Neurons with graded response have collective… (More)

- Max H. Garzon, Fernanda Botelho
- Neurocomputing
- 1999

Autism spectrum disorders (ASDs) are a triad of disturbances affecting the areas of communication, social interaction and behavior. In educational contexts, without appropriate intervention methodologies, these limitations can be deeply disabling. Our research promotes the communicative competence of children with ASDs It extends the current… (More)

- Max H. Garzon, Fernanda Botelho
- NIPS
- 1993

We prove that except possibly for small exceptional sets, discrete-time analog neural nets are globally observable, i.e. all their corrupted pseudo-orbits on computer simulations actually reflect the true dynamical behavior of the network. Locally finite discrete (boolean) neural networks are observable without exception.

- Fernanda Botelho, James Jamison, Angela Murdock
- 2006

We study the existence and stability of pulse stationary solutions of an integro-differential equation modelling the coarse-grained averaged activity of a single layer of interconnected neurons. The neuronal connections considered are laterally oscilla-tory with an exponential rate of decay and variable phase. We identify regions in the parameter space… (More)

- Fernanda Botelho
- Neural Networks
- 1999

- Fernanda Botelho, Max H. Garzon
- Theor. Comput. Sci.
- 1994

- J. Angela Hart Murdock, Fernanda Botelho, ANGELA HART MURDOCK
- 2007

We study the existence and stability of stationary solutions of an integro-differential equation modelling the activity of a single layer of interconnected neurons. The multi-parameter neural connections considered incorporate lateral oscillations with an exponential rate of decay. We identify regions in the parameter space where solutions exhibit areas of… (More)

- Fernanda Botelho
- 1998

We consider discrete and recurrent neural network models with saturated linear neurons. A denition of capacity is discussed and conditions that assure innite capacity are established. The aim of this paper is to study networks with maximum capacity and to what extend maximal capacity relates to the network connecting weights.

We present a relation between the rotation of chain transitive sets and the rotation shadowing for annulus homeomorphisms isotopic to identity.