Description:AS ATTACHED , Claims:I/We Claim:
1. Data stream (45) having a representation of a neural network (10) encoded
thereinto, the data stream (45) comprising a numerical computation
5 representation parameter (120) indicating a numerical representation and bit
size at which neural network parameters (32) of the neural network, which
are encoded into the data stream, are to be represented when using the neural
network (10) for inference.
10 2. Apparatus for encoding a representation of a neural network (10) into a data
stream (45), wherein the apparatus is configured to provide the data stream
(45) with a numerical computation representation parameter (120)
indicating a numerical representation and bit size at which neural network
parameters (32) of the neural network, which are encoded into the data
15 stream (45), are to be represented when using the neural network (10) for
inference.
3. Apparatus for decoding a representation of a neural network (10) from a
data stream (45), wherein the apparatus is configured to decode from the
20 data stream (45) a numerical computation representation parameter (120)
indicating a numerical representation and bit size at which neural network
parameters (32) of the neural network, which are encoded into the data
stream (45), are to be represented when using the neural network (10) for
inference, and to use the numerical representation and bit size for
25 representing the neural network parameters (32) decoded from the data
stream (45).
4. Apparatus of claim 3, wherein the data stream (45), is structured into
individually accessible sub-portions (43, 44, 240), each individually
30 accessible sub-portion representing a corresponding neural network portion
of the neural network, so that each individually accessible sub-portion is
completely traversed by the coding order (104) before a subsequent
individually accessible sub-portion is traversed by the coding order (104),
wherein the apparatus is configured to decode, from the data stream (45),
for a predetermined individually accessible sub-portion the neural network
5 parameter and a type parameter indicting a parameter type of the neural
network parameter decoded from the predetermined individually accessible
sub-portion.
5. Apparatus of claim 4, wherein the type parameter discriminates, at least,
10 between neural network weights and neural network biases.
6. Apparatus of any of the previous claims 3 to 5, wherein the data stream (45),
is structured into one or more individually accessible portions (200), each
one or more individually accessible portion representing a corresponding
15 neural network layer (210, 30) of the neural network, and
wherein the apparatus is configured to decode, from the data stream
(45), for a predetermined neural network layer, a neural network layer type
parameter (130) indicating a neural network layer type of the predetermined
neural network layer of the neural network.
20
7. Apparatus of claim 6, wherein the neural network layer type parameter (130)
discriminates, at least, between a fully-connected and a convolutional layer
type.
25 8. Apparatus of any of claims 3 to 7, wherein the apparatus is configured to
decode a representation of a neural network (10) from the data stream (45),
wherein the data stream (45) is structured into one or more individually
accessible portions (200), each individually accessible portion representing
a corresponding neural network layer (210, 30) of the neural network, and
30 wherein the data stream (45) is, within a predetermined portion, further
structured into individually accessible sub-portions (43, 44, 240), each sub-
portion (43, 44, 240) representing a corresponding neural network portion
of the respective neural network layer (210, 30) of the neural network,
wherein the apparatus is configured to decode from the data stream (45), for
each of one or more predetermined individually accessible sub-portions (43,
5 44, 240)
a start code (242) at which the respective predetermined individually
accessible sub-portion begins, and/or
a pointer (244) pointing to a beginning of the respective predetermined
individually accessible sub-portion, and/or
10 a data stream length parameter indicating a data stream length (246) of the
respective predetermined individually accessible sub-portion for skipping
the respective predetermined individually accessible sub-portion in parsing
the data stream (45).
15 9. Apparatus of claim 8, wherein the apparatus is configured to decode, from
the data stream (45), the representation of the neural network using context-
adaptive arithmetic decoding and using context initialization at a start of
each individually accessible portion and each individually accessible sub-
portion.
20
10. Apparatus of any previous claim 3 to 9, wherein the apparatus is configured
to decode a representation of a neural network (10) from a data stream (45),
wherein the data stream (45) is structured into individually accessible
portions (200), each portion representing a corresponding neural network
25 portion of the neural network, wherein the apparatus is configured to decode
from the data stream (45), for each of one or more predetermined
individually accessible portions, an identification parameter (310) for
identifying the respective predetermined individually accessible portion.
11. Apparatus of claim 10, wherein the identification parameter (310) is related
to the respective predetermined individually accessible portion via a hash
function or error detection code or error correction code.
5 12. Apparatus of claim 10 or claim 11, wherein the apparatus is configured to
decode, from the data stream (45), a higher-level identification parameter
(310) for identifying a collection of more than one predetermined
individually accessible portion.
10 13. Apparatus of claim 12, wherein the higher-level identification parameter
(310) is related to the identification parameters (310) of the more than one
predetermined individually accessible portion via a hash function or error
detection code or error correction code.
15 14. Apparatus of any previous claim 3 to 13, wherein the apparatus is configured
to decode a representation of a neural network (10) from a data stream (45),
wherein the data stream (45) is structured into individually accessible
portions (200), each portion representing a corresponding neural network
portion of the neural network, wherein the apparatus is configured to decode
20 from the data stream (45), for each of one or more predetermined
individually accessible portions a supplemental data (350) for
supplementing the representation of the neural network.
15. Apparatus of claim 14, wherein the data stream (45) indicates the
25 supplemental data (350) as being dispensable for inference based on the
neural network.
16. Apparatus of claim 14 or claim 15, wherein the apparatus is configured to
decode the supplemental data (350) for supplementing the representation of
30 the neural network for the one or more predetermined individually
accessible portions (200) from further individually accessible portions,
wherein the data stream (45) comprises for each of the one or more
predetermined individually accessible portions a corresponding further
predetermined individually accessible portion relating to the neural network
portion to which the respective predetermined individually accessible
5 portion corresponds.
17. Apparatus of any previous claim 14 to 16, wherein the supplemental data
(350) relates to
10 relevance scores of neural network parameters (32), and/or
perturbation robustness of neural network parameters (32).
18. Apparatus of any previous claim 3 to 17, for decoding a representation of a
neural network (10) from a data stream (45), wherein the apparatus is
15 configured to decode from the data stream (45) hierarchical control data
(400) structured into a sequence (410) of control data portions (420),
wherein the control data portions provide information on the neural network
at increasing details along the sequence of control data portions.
20 19. Apparatus of claim 18, wherein at least some of the control data portions
(420) provide information on the neural network which is partially
redundant.
20. Apparatus of claim 18 or claim 19, wherein a first control data portion
25 provides the information on the neural network by way of indicating a
default neural network type implying default settings and a second control
data portion comprises a parameter to indicate each of the default settings.
21. Apparatus for performing an inference using a neural network, comprising
an apparatus for decoding a data stream (45) according to any of claims 3 to
5 20, so as to derive from the data stream (45) the neural network, and
a processor configured to perform the inference based on the neural network.
22. Method for encoding a representation of a neural network into a data stream,
10 providing the data stream with a numerical computation representation
parameter indicating a numerical representation and bit size at which neural
network parameters of the neural network, which are encoded into the data
stream, are to be represented when using the neural network for inference.
15 23. Method for decoding a representation of a neural network from a data
stream, wherein the method comprises decoding from the data stream a
numerical computation representation parameter indicating a numerical
representation and bit size at which neural network parameters of the neural
network, which are encoded into the data stream, are to be represented when
20 using the neural network for inference, and to use the numerical
representation and bit size for representing the neural network parameters
decoded from the data stream.
24. Computer program for, when executed by a computer, causing the computer
25 to perform the method of claim 22 or claim 23.