Description:AS ATTACHED , Claims:I/We Claim:
1. Data stream (45) having a representation of a neural network (10) encoded
thereinto, wherein the data stream (45) is structured into one or more
individually accessible 5 portions (200), each individually accessible portion
representing a corresponding neural network layer (210, 30) of the neural
network, 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
10 of the respective neural network layer (210, 30) of the neural network,
wherein the data stream (45) comprises for each of one or more predetermined
individually accessible sub-portions (43, 44, 240)
a start code (242) at which the respective predetermined individually
accessible sub-portion begins, and/or
15 a pointer (244) pointing to a beginning of the respective
predetermined individually accessible sub-portion, and/or
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
20 sub-portion in parsing the data stream (45).
2. Apparatus for encoding a representation of a neural network (10) into a data
stream (45), so that the data stream (45) is structured into one or more
individually accessible portions (200), each individually accessible portion
25 representing a corresponding neural network layer (210, 30) of the neural
network, and so that 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 of the neural network, wherein
30 the apparatus is configured to provide the data stream (45) with, for each of
62
one or more predetermined individually accessible sub-portions (43, 44,
240)
a start code (242) at which the respective predetermined individually
accessible sub-portion begins, and/or
a pointer (244) 5 pointing to a beginning of the respective
predetermined individually accessible sub-portion, and/or
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
10 sub-portion in parsing the data stream (45).
3. Apparatus for decoding a representation of a neural network (10) from a data
stream (45), wherein the data stream (45) is structured into one or more
individually accessible portions (200), each individually accessible portion
15 representing a corresponding neural network layer (210, 30) of the neural
network, and 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
20 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, 44, 240)
a start code (242) at which the respective predetermined individually
accessible sub- portion begins, and/or
25 a pointer (244) pointing to a beginning of the respective
predetermined individually accessible sub-portion, and/or
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
30 sub-portion in parsing the data stream (45).
63
4. Apparatus of claim 3, wherein the apparatus is configured to decode, from the
data stream (45), the representation of the neural network using contextadaptive
arithmetic decoding and using context initialization at a start of each
individually accessible portion and each individually accessible sub-portion.
5
5. Apparatus of claim 3 or claim 4, 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 individually accessible portion representing a corresponding
10 neural network 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 (200), a processing option
parameter (250) indicating one or more processing options (252) which have
to be used or which may optionally be used when using the neural network
15 (10) for inference.
6. Apparatus of claim 5, wherein the processing option parameter (250)
indicates the one or more available processing options (252) out of a set of
predetermined processing options (252) including
20 parallel processing capability of the respective predetermined
individually accessible portion; and/or
sample wise parallel processing capability (2522) of the respective
predetermined individually accessible portion; and/or
channel wise parallel processing capability (2521) of the respective
25 predetermined individually accessible portion; and/or
classification category wise parallel processing capability of the
respective predetermined individually accessible portion; and/or
dependency of the neural network portion represented by the
respective predetermined individually accessible portion on a
30 computation result gained from another individually accessibly
portion of the data stream (45) relating to the same neural network
64
portion but belonging to another version of versions (330) of the
neural network which are encoded into the data stream (45) in a
layered manner.
7. Apparatus of any previous claim 3 5 to 6, 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
10 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.
8. Apparatus of claim 7, wherein the identification parameter (310) is related
15 to the respective predetermined individually accessible portion via a hash
function or error detection code or error correction code.
9. Apparatus of claim 7 or claim 8, wherein the apparatus is configured to
decode, from the data stream (45), a higher-level identification parameter
20 (310) for identifying a collection of more than one predetermined
individually accessible portion.
10. Apparatus of claim 9, wherein the higher-level identification parameter
(310) is related to the identification parameters (310) of the more than one
25 predetermined individually accessible portion via a hash function or error
detection code or error correction code.
11. Apparatus of any previous claim 3 to 10, wherein the apparatus is configured
to decode a representation of a neural network (10) from a data stream (45),
30 wherein the data stream (45) is structured into individually accessible
portions (200), each portion representing a corresponding neural network
65
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 a supplemental data (350) for
supplementing the representation of the neural network.
5
12. Apparatus of claim 11, wherein the data stream (45) indicates the
supplemental data (350) as being dispensable for inference based on the
neural network.
10 13. Apparatus of claim 11 or claim 12, wherein the apparatus is configured to
decode the supplemental data (350) for supplementing the representation of
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
15 predetermined individually accessible portions a corresponding further
predetermined individually accessible portion relating to the neural network
portion to which the respective predetermined individually accessible
portion corresponds.
20 14. Apparatus of any previous claim 11 to 13, wherein the supplemental data
(350) relates to
relevance scores of neural network parameters (32), and/or
perturbation robustness of neural network parameters (32).
25
15. Apparatus of any previous claim 3 to 14, for decoding a representation of a
neural network (10) from a data stream (45), wherein the apparatus is
configured to decode from the data stream (45) hierarchical control data
(400) structured into a sequence (410) of control data portions (420),
30 wherein the control data portions provide information on the neural network
at increasing details along the sequence of control data portions.
66
16. Apparatus of claim 15, wherein at least some of the control data portions
(420) provide information on the neural network which is partially
redundant.
5
17. Apparatus of claim 15 or claim 16, wherein a first control data portion
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.
10
18. 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
17, so as to derive from the data stream (45) the neural network, and
15
a processor configured to perform the inference based on the neural network.
19. Method for encoding a representation of a neural network into a data stream,
so that the data stream is structured into one or more individually accessible
20 portions, each individually accessible portion representing a corresponding
neural network layer of the neural network, and so that the data stream is,
within a predetermined portion, further structured into individually
accessible sub-portions, each sub-portion representing a corresponding
neural network portion of the respective neural network layer of the neural
25 network, wherein the method comprises providing the data stream with, for
each of one or more predetermined individually accessible sub-portions
a start code at which the respective predetermined individually
accessible sub-portion begins, and/or
a pointer pointing to a beginning of the respective predetermined
30 individually accessible sub-portion, and/or
67
a data stream length parameter indicating a data stream length of the
respective predetermined individually accessible sub-portion for
skipping the respective predetermined individually accessible subportion
in parsing the data stream.
5
20. Method for decoding a representation of a neural network from a data
stream, wherein the data stream is structured into one or more individually
accessible portions, each individually accessible portion representing a
corresponding neural network layer of the neural network, and wherein the
10 data stream is, within a predetermined portion, further structured into
individually accessible sub-portions, each sub-portion representing a
corresponding neural network portion of the respective neural network layer
of the neural network, wherein the method comprises decoding from the data
stream, for each of one or more predetermined individually accessible sub15
portions
a start code at which the respective predetermined individually
accessible sub-portion begins, and/or
a pointer pointing to a beginning of the respective predetermined
individually accessible sub-portion, and/or
20 a data stream length parameter indicating a data stream length of the
respective predetermined individually accessible sub-portion for
skipping the respective predetermined individually accessible subportion
in parsing the data stream.
25 21. Computer program for, when executed by a computer, causing the computer
to perform the method of claim 19 or claim 20.