Types used in Cerebro

Clear Types

Clear types are used to store values that are visible to every party. Use these values to store global variables and constants which do not depend on any user’s input.

c_int: A clear integer modulo p, where p is the prime field. :

x = c_int(5)

Main operations supported with clear integers

• Addition

• Subtraction

• Multiplication

• Comparisons

• Shifting

• Bit-decompose

• Hash digest

c_fix: A clear fixed-point number modulo p. :

x = c_fix(2.5)

The precision must be specified at the top of the .mpc program as follows: Params.set_params(k=, f=), where k indicates the number of bits used to represent a floating point number and f represents the number of bits to represent the decimal places.

Main operations supported with clear fixed-point values

• Addition

• Subtraction

• Multiplication

• Comparisons

• Bit-decompose

• Division

Additionally, Cerebro supports containers for clear integers and clear fixed-point values.

c_int_mat: A clear int matrix. :

# mat is assigned to a 5 x 7 clear integer matrix
mat = c_int_mat(5, 7)

c_fix_mat: A clear fixed-pointed matrix. :

# mat is assigned to a 5x7 clear fixed-point matrix
mat = c_fix_mat(5, 7)

List of operations supported with clear-valued containers

The list of operations that be done on matrices can be found in the Cerebreo mllib section.

Secret Types

Secret types are used to store values that are not visible to any party. Use these values to store values that depend on individual party’s inputs.

s_int: A clear integer modulo p, where p is the prime field. :

x = s_int(5)

List of operations supported with secret values

• Addition

• Subtraction

• Multiplication

• Comparisons

• Shifting

• Bit-decompose

s_fix: A clear fixed-point number modulo p. :

x = s_fix(2.5)

The precision must be specified at the top of the .mpc program as follows: Params.set_params(k=, f=), where k indicates the number of bits used to represent a floating point number and f represents the number of bits to represent the decimal places.

List of operations supported with secret fixed-point values

• Addition

• Subtraction

• Multiplication

• Comparisons

• Division

Additionally, Cerebro supports containers for clear integers and clear fixed-point values.

s_int_mat: A clear int matrix. :

# mat is assigned to a 5 x 7 secret integer matrix
mat = s_int_mat(5, 7)

s_fix_mat: A clear fixed-pointed matrix. :

# mat is assigned to a 5x7 secret fixed-point matrix
mat = s_fix_mat(5, 7)

List of operations supported with secret-valued containers

The list of operations that be done on matrices can be found in the Cerebreo mllib section.